Methods for Producing Multimeric Proteins in Eukaryotic Host Cells

Abstract

The present disclosure relates to the production of multimeric polypeptides (e.g., those comprising two or more subunits, each subunit comprising two or more polypeptide chains, e.g., multispecific or bispecific antibodies) in eukaryotic host cells. In some embodiments, provided herein are methods for producing multimeric polypeptides in eukaryotic host cells using host cells comprising a polynucleotide comprising a translation initiation sequence operably linked to an open reading frame encoding each polypeptide of the multimeric polypeptide. Advantageously, the present disclosure demonstrates that by adjusting the strength of the translation initiation sequence linked to each open reading frame, mispaired by-products can be reduced and the yield of multimeric polypeptides can be increased. The present disclosure further provides cells, screening methods, and kits related thereto.

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application No. 62 / 796,014, filed January 23, 2019, the entirety of which is incorporated by reference herein.

[0002] Submitting a sequence listing as an ASCII text file

[0002] The contents of the following ASCII text file submission are incorporated by reference in their entirety into this specification: Computer Readable Form (CRF) of Sequence Listing (Filename: 146392045240SEQLIST.TXT, Recorded: Jan. 22, 2020, Size: 5KB).

[0003]

[0003] The present disclosure relates to methods for producing multimeric polypeptides in eukaryotic (eg, mammalian) host cells, and related cells, methods, and kits or articles of manufacture thereof. [Background technology]

[0004]

[0004] Therapeutic antibodies represent the most successful biological drugs. Over the past 30 years, more than 40 therapeutic antibodies have been approved for clinical use in a variety of indications, including cancer, autoimmunity, infectious diseases, and vascular diseases. Thus, therapeutic antibodies are one of the fastest growing areas in the pharmaceutical industry, and their clinical impact is significant.

[0005]

[0005] Most diseases have multiple parallel signaling pathways, so multiple inhibition of receptors and ligands may provide a superior therapeutic effect. Thus, bispecific antibodies (bsAbs), capable of binding two different epitopes, have potential applications in the treatment of, among others, inflammatory and autoimmune diseases (Chan AC, Carter PJ. Nat Rev Immunol. 2010;10(5):301-16), cancer (Kou G, Shi J, Chen L, Zhang D, Hou S, Zhao L, et al. Cancer Lett. 2010;299(2):130-6;Dong J, Sereno A, Aivazian D, Langley E, Miller BR, Snyder WB, et al. MAbs. 2011;3(3):273-88), and infectious diseases (De Bernardis F, Liu H, O'Mahony R, La Valle R, Bartollino S, Sandini S, et al. J Infect Dis. 2007;195(1):149-57;Laventie BJ, Rademaker HJ, Saleh M, de Boer E, Janssens R, Bourcier T, et al. Proc Natl Acad Sci US A. 2011;108(39):16404-9) and have emerged as promising therapeutic applications for a variety of diseases (Kontermann RE. MAbs. 2012;4(2):182-97;Brinkmann U, Kontermann RE. MAbs. 2017;9(2):182-212;Carter PJ, Lazar GA. Nat Rev Drug Discov. 2018;17(3):197-223). Over the past few decades, genetic engineering has yielded over 60 different bispecific antibody formats, improving the immunogenicity, pharmacokinetic properties, and distribution of bsAbs (Spiess C, Zhai Q, Carter PJ. Mol Immunol. 2015;67(2 Pt A):95-106).

[0006]

[0006] Regardless of the use of bsAbs, the manufacturing process has been challenging as two different heavy chains are promiscuous with two different light chains resulting in 16 different combinations, only one of which is bispecific, making it difficult to produce sufficient quantity, quality, and assembly to support preclinical and clinical development. Antibody engineering has been used to alleviate the problem of chain mispairing by heterodimerizing heavy or light chains with knob-into-hole and crossMab technologies, respectively (Ridgway JB, Presta LG, Carter P. Protein Eng. 1996;9(7):617-21; Atwell S, Ridgway JB, Wells JA, Carter P. J Mol Biol. 1997;270(1):26-35; Merchant AM, Zhu Z, Yuan JQ, Goddard A, Adams CW, Presta LG, et al. Nat Biotechnol. 1998;16(7):677-81). A solution to the manufacturing process employs separate expression of half antibodies and their subsequent in vitro assembly into bsAb (Spiess C, Merchant M, Huang A, Zheng Z, Yang NY, Peng J, et al. Nat Biotechnol. 2013;31(8):753-8), however this is associated with a time-consuming and costly manufacturing process due to the generation of two different stable cell lines.

[0007]

[0007] Thus, co-expressing two antibodies in a single cell may be easier, but at the same time, more extensive optimization of the expression system is often required to direct the light chain to its cognate heavy chain, avoiding the presence of undesirable by-products and the need for complex purification processes, and to achieve adequate amounts of correctly assembled bsAb.

[0008]

[0008] Over the past few decades, mammalian cells, such as Chinese Hamster Ovary (CHO) cells, have emerged as the primary host for the biopharmaceutical industry (Wurm FM. Nat Biotechnol. 2004;22(11):1393-8). Bioprocess innovations and cell engineering efforts have improved product titers (Ayyar BV, Arora S, Ravi SS. Methods. 2017;116:51-62;Kelley B, Kiss R, Laird M. Adv Biochem Eng Biotechnol. 2018. Epub 2018 / 05 / 04). However, poorly understood cellular processes and gene control mechanisms remain obstacles to cell growth, specific productivity, and protein quality. Currently, there are few systematic approaches to precisely control recombinant protein translation levels in mammalian cells.

[0009]

[0009] Although advances have been made in the engineering of complex antibody formats such as bispecific antibodies, manufacturability in a single mammalian expression system often shows poor performance. Low titers and poor product quality are two of the factors that make stable production of bsAbs difficult. As a result, there is an urgent need to identify the limiting steps in the production system and provide methods that allow improving the production of correctly assembled multimeric polypeptides, for example in eukaryotic or mammalian host cell systems.

[0010]

[0010] All references cited in this specification, including patent applications, patent publications, non-patent literature, and UniProtKB / Swiss-Prot accession numbers, are incorporated by reference in their entirety into this specification, as if each individual reference was specifically and individually indicated to be incorporated by reference. Summary of the Invention

[0011]

[0011] To meet these and other needs, methods are provided herein for producing multimeric polypeptides in eukaryotic (e.g., mammalian) host cells. These multimeric polypeptides comprise two or more subunits, each of which comprises two or more polypeptide chains (e.g., similar to multispecific or bispecific antibodies). Advantageously, these methods allow for more precise manipulation of the translation level of each polypeptide chain in the multimeric polypeptide, improving the production of correctly assembled multimeric polypeptides.

[0012]

[0012] Certain aspects of the present disclosure relate to methods for producing multimeric polypeptides in eukaryotic host cells. In some embodiments, the multimeric polypeptide comprises a first subunit comprising a first polypeptide chain and a second polypeptide chain, and a second subunit comprising a third polypeptide chain and a fourth polypeptide chain. In some embodiments, the method comprises providing a eukaryotic host cell, culturing the eukaryotic host cell under conditions suitable for expression of the first, second, third, and fourth polypeptide chains, wherein upon expression, the first, second, third, and fourth polypeptide chains form a multimeric polypeptide, and recovering the multimeric polypeptide produced by the eukaryotic host cell. In some embodiments, the eukaryotic host cell comprises a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding a first polypeptide chain, a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding a second polypeptide chain, a third polynucleotide comprising a third translation initiation sequence operably linked to a third open reading frame encoding a third polypeptide chain, and a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to a fourth open reading frame encoding a fourth polypeptide chain. In some embodiments, the first subunit is expressed at a lower level than the second subunit when each subunit is expressed individually in a eukaryotic host cell, and one or both of the first translation initiation sequence and the second translation initiation sequence are weaker than one or both of the third translation initiation sequence and the fourth translation initiation sequence. In some embodiments, the first subunit is expressed at a lower level than the second subunit when all subunits are expressed together in the same eukaryotic host cell, and one or both of the first translation initiation sequence and the second translation initiation sequence are weaker than one or both of the third translation initiation sequence and the fourth translation initiation sequence. In some embodiments, the multimeric polypeptide is a bispecific antibody.In some embodiments, the multimeric polypeptide is a bispecific antibody, wherein the first and third polypeptide chains are antibody heavy chains, the second and fourth polypeptide chains are antibody light chains, the first subunit is a first half antibody that binds to a first antigen, and the second subunit is a second half antibody that binds to a second antigen.

[0013]

[0013] Another aspect of the present disclosure is a method for producing a bispecific antibody in a eukaryotic host cell, the bispecific antibody comprising a first half-antibody comprising a first antibody heavy chain and a first antibody light chain, and a second half-antibody comprising a second antibody heavy chain and a second antibody light chain, comprising: (a) providing a eukaryotic host cell, the eukaryotic host cell being capable of detecting a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding the first antibody heavy chain, a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding the first antibody light chain, a third polynucleotide comprising a third translation initiation sequence operably linked to a third open reading frame encoding the second antibody heavy chain, and a fourth open reading frame encoding the second antibody light chain. and a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to the third translation initiation sequence and a fourth translation initiation sequence operably linked to the fourth translation initiation sequence, wherein when each half antibody is expressed individually in the eukaryotic host cell, the first half antibody is expressed at a lower level than the second half antibody and one or both of the first translation initiation sequence and the second translation initiation sequence is weaker than one or both of the third translation initiation sequence and the fourth translation initiation sequence; (b) culturing the eukaryotic host cell under conditions suitable for expression of the first and second antibody heavy and light chains, wherein the first and second antibody heavy and light chains form a bispecific antibody, and wherein the first half antibody binds to the first antigen and the second half antibody binds to the second antigen; and (c) recovering the bispecific antibody produced by the eukaryotic host cell.

[0014]

[0014] In some embodiments, the first antibody heavy chain comprises a first antibody Fc region comprising a CH2 domain and a CH3 domain, and the second antibody heavy chain comprises a second antibody Fc region comprising a CH2 domain and a CH3 domain, and the CH3 domain of the first antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a smaller side chain volume, thereby generating a hole on the surface of the CH3 domain of the first antibody Fc region that interacts with the CH3 domain of the second antibody Fc region, and the CH3 domain of the second antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a larger side chain volume, thereby generating a knob on the surface of the CH3 domain of the second antibody Fc region that interacts with the CH3 domain of the first antibody Fc region. In some embodiments, the first antibody heavy chain comprises a first antibody Fc region comprising a CH2 domain and a CH3 domain, wherein the second antibody heavy chain comprises a second antibody Fc region comprising a CH2 domain and a CH3 domain, and the CH3 domain of the second antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a smaller side chain volume, thereby generating a hole on the surface of the CH3 domain of the second antibody Fc region that interacts with the CH3 domain of the first antibody Fc region, and the CH3 domain of the first antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a larger side chain volume, thereby generating a knob on the surface of the CH3 domain of the first antibody Fc region that interacts with the CH3 domain of the second antibody Fc region. In some embodiments, the knob mutation comprises at least one of T366Y, T366W, T394W, and F405W, numbered based on human IgG1 according to the EU index. In some embodiments, the whole mutations include at least one of F405A, Y407T, Y407A, T366S, L368A, Y407V, and T394S, numbered based on the EU index for human IgG1.In some embodiments, the knob mutation comprises T366W, where the hole mutation comprises at least one, at least two, or all three of T366S, L368A, and Y407V, numbered according to the EU index for human IgG1. In some embodiments, the first antibody light chain comprises a first mutation and the first antibody heavy chain comprises a second mutation, where the first and second mutations promote selective association of the first antibody light chain with the first antibody heavy chain. In some embodiments, the first mutation comprises an amino acid substitution at V133 and / or the second mutation comprises an amino acid substitution at S183 (numbering according to the EU index). In some embodiments, the S183 substitution is selected from the group consisting of S183A, S183T, S183V, S183Y, S183F, S183H, S183N, S183D, S183E, S183R, and S183K, and / or the V133 substitution is selected from the group consisting of V133E, V133S, V133L, V133W, V133K, V133R, and V133D. In some embodiments, the second antibody light chain comprises a third mutation and the second antibody heavy chain comprises a fourth mutation, and the third and fourth mutations promote selective association of the second antibody light chain with the second antibody heavy chain. In some embodiments, the third mutation comprises an amino acid substitution at V133 and / or the fourth mutation comprises an amino acid substitution at S183 (numbering according to the EU index). In some embodiments, the S183 substitution is selected from the group consisting of S183A, S183T, S183V, S183Y, S183F, S183H, S183N, S183D, S183E, S183R, and S183K, and / or the V133 substitution is selected from the group consisting of V133E, V133S, V133L, V133W, V133K, V133R, and V133D. In some embodiments, the amino acid substitution at S183 results in a positively charged residue (e.g., S183K) and the amino acid substitution at V133 results in a negatively charged residue (e.g., V133E). In some embodiments, the amino acid substitution at S183 results in a negatively charged residue (eg, S183E) and the amino acid substitution at V133 results in a positively charged residue (eg, V133K).

[0015] In some embodiments, the method produces more multimeric polypeptides in eukaryotic host cells, for example, compared to production in which one or more, two or more, three or more, or four of the polynucleotides encoding the polypeptide chains of the multimeric polypeptides comprise an open reading frame operably linked to a natural or unmodified translation initiation sequence, or each of the polynucleotides encoding the polypeptide chains of the multimeric polypeptides comprises the same translation initiation sequence. In some embodiments, the method produces fewer mismatched by-products in eukaryotic host cells, for example, compared to production in which one or more, two or more, three or more, or four of the polynucleotides encoding the polypeptide chains of the multimeric polypeptides comprise an open reading frame operably linked to a natural or unmodified translation initiation sequence, or each of the polynucleotides encoding the polypeptide chains of the multimeric polypeptides comprises the same translation initiation sequence. For example, transient or stable transfectants can be cultured and production of multimeric polypeptides can be measured. Fed-batch or perfusion culture can be performed and the titer of the product in the cell medium or on the cell surface can be measured. For surface expression, the cells can be stained with antibodies to detect the product and analyzed, for example, by flow cytometry. The quality and / or purity of the product can be evaluated, for example, using electrophoresis and / or mass spectrometry. In some embodiments, when each subunit is expressed individually in a eukaryotic host cell, one or both polypeptide chains of the first subunit are translated at a slower rate than one or both polypeptide chains of the second subunit. In some embodiments, when each subunit is expressed individually in a eukaryotic host cell, one or both polypeptide chains of the first subunit are folded more slowly and / or less efficiently than one or both polypeptide chains of the second subunit. In some embodiments, when each subunit is expressed individually in a eukaryotic host cell, the first subunit is assembled at a slower rate than the second subunit.In some embodiments, the first and / or second translation initiation sequence is at least 5%, at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, 5% to 30%, 5% to 50%, 5% to 75%, 10% to 30%, 10% to 50%, 10% to 75%, 25% to 50%, 25% to 75%, 25% to 100%, 50% to 75%, 50% to 100%, or 75% to 100% weaker than the third and / or fourth translation initiation sequence. In some embodiments, the first and / or second translation initiation sequence is at least 1.3 times, at least 1.5 times, at least 2 times, at least 2.5 times, at least 3 times, at least 3.5 times, at least 4 times, at least 4.5 times, at least 5 times, at least 6 times, at least 7 times, at least 8 times, at least 9 times, at least 10 times, 1.3 times to 3 times, 1.5 times to 3 times, 2 times to 10 times, 2 times to 5 times, 3 times to 5 times, 3 times to 10 times, 5 times to 10 times, or 7 times to 10 times weaker than the third and / or fourth translation initiation sequence. In some embodiments, a first subunit or half antibody is expressed at a level that is at least 5%, at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, 5% to 30%, 5% to 50%, 5% to 75%, 10% to 30%, 10% to 50%, 10% to 75%, 25% to 50%, 25% to 75%, 25% to 100%, 50% to 75%, 50% to 100%, or 75% to 100% lower than the expression level of a second subunit or half antibody, for example when each subunit or half antibody is expressed individually in a eukaryotic host cell.In some embodiments, the first subunit or half antibody is expressed at a level that is at least 1.3-fold, at least 1.5-fold, at least 2-fold, at least 2.5-fold, at least 3-fold, at least 3.5-fold, at least 4-fold, at least 4.5-fold, at least 5-fold, at least 6-fold, at least 7-fold, at least 8-fold, at least 9-fold, at least 10-fold, 1.3-fold to 3-fold, 1.5-fold to 3-fold, 2-fold to 10-fold, 2-fold to 5-fold, 3-fold to 5-fold, 3-fold to 10-fold, 5-fold to 10-fold, or 7-fold to 10-fold lower than the expression level of the second subunit or half antibody, for example, when each subunit or half antibody is expressed individually in a eukaryotic host cell. In some embodiments, the first subunit or half antibody is expressed at a level that is 0.2- to 0.8-fold, less than 0.8-fold, less than 0.5-fold, or less than 0.3-fold greater than the expression level of the second subunit or half antibody, for example, when each subunit or half antibody is expressed individually in a eukaryotic host cell.

[0016]

[0016] Another aspect of the present disclosure relates to a plurality of compositions of multimeric polypeptides or bispecific antibodies, wherein each multimeric polypeptide or bispecific antibody of the plurality of compositions is produced by a method according to any one of the embodiments described herein.

[0017] Another aspect of the present disclosure is a recombinant eukaryotic host cell for expression of a non-naturally occurring multimeric polypeptide comprising a first subunit comprising a first polypeptide chain and a second polypeptide chain, and a second subunit comprising a third polypeptide chain and a fourth polypeptide chain, wherein the host cell is configured to express a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding the first polypeptide chain, a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding the second polypeptide chain, and a third open reading frame encoding the third polypeptide chain. and a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to a fourth open reading frame encoding a fourth polypeptide chain, wherein the first subunit is expressed at a lower level than the second subunit and one or both of the first translation initiation sequence and the second translation initiation sequence are weaker than one or both of the third translation initiation sequence and the fourth translation initiation sequence when each subunit is expressed individually in the recombinant eukaryotic host cell or when all the subunits are expressed together in the same eukaryotic host cell. In some embodiments, the multimeric polypeptide is a bispecific antibody, wherein the first and third polypeptide chains are antibody heavy chains and the second and fourth polypeptide chains are antibody light chains, and the first subunit is a first half antibody that binds a first antigen and the second subunit is a second half antibody that binds a second antigen.

[0018]

[0018] Another aspect of the present disclosure is a recombinant eukaryotic host cell for expression of a bispecific antibody comprising a first half-antibody comprising a first antibody heavy chain and a first antibody light chain, and a second half-antibody comprising a second antibody heavy chain and a second antibody light chain, the host cell comprising: (a) a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding the first antibody heavy chain; (b) a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding the first antibody light chain; and (c) a third open reading frame encoding the second antibody heavy chain. and (d) a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to a fourth open reading frame encoding a second antibody light chain, wherein when each half antibody is expressed individually in a recombinant eukaryotic host cell, the first half antibody is expressed at a lower level than the second half antibody, and one or both of the first translation initiation sequence and the second translation initiation sequence are weaker than one or both of the third translation initiation sequence and the fourth translation initiation sequence. In some embodiments, the first and second antibody heavy and light chains form a bispecific antibody, wherein the first half antibody binds a first antigen and the second half antibody binds a second antigen.

[0019]

[0019] In some embodiments, the first antibody heavy chain comprises a first antibody Fc region comprising a CH2 domain and a CH3 domain, and the second antibody heavy chain comprises a second antibody Fc region comprising a CH2 domain and a CH3 domain, and the CH3 domain of the first antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a smaller side chain volume, thereby generating a hole on the surface of the CH3 domain of the first antibody Fc region that interacts with the CH3 domain of the second antibody Fc region, and the CH3 domain of the second antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a larger side chain volume, thereby generating a knob on the surface of the CH3 domain of the second antibody Fc region that interacts with the CH3 domain of the first antibody Fc region. In some embodiments, the first antibody heavy chain comprises a first antibody Fc region comprising a CH2 domain and a CH3 domain, wherein the second antibody heavy chain comprises a second antibody Fc region comprising a CH2 domain and a CH3 domain, and the CH3 domain of the second antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a smaller side chain volume, thereby generating a hole on the surface of the CH3 domain of the second antibody Fc region that interacts with the CH3 domain of the first antibody Fc region, and the CH3 domain of the first antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a larger side chain volume, thereby generating a knob on the surface of the CH3 domain of the first antibody Fc region that interacts with the CH3 domain of the second antibody Fc region. In some embodiments, the knob mutation comprises at least one of T366Y, T366W, T394W, and F405W, numbered based on human IgG1 according to the EU index. In some embodiments, the whole mutations include at least one of F405A, Y407T, Y407A, T366S, L368A, Y407V, and T394S, numbered based on the EU index for human IgG1.In some embodiments, the knob mutation comprises T366W, where the hole mutation comprises at least one, at least two, or all three of T366S, L368A, and Y407V, numbered according to the EU index for human IgG1. In some embodiments, the first antibody light chain comprises a first mutation and the first antibody heavy chain comprises a second mutation, where the first and second mutations promote selective association of the first antibody light chain with the first antibody heavy chain. In some embodiments, the first mutation comprises an amino acid substitution at V133 and / or the second mutation comprises an amino acid substitution at S183 (numbering according to the EU index). In some embodiments, the S183 substitution is selected from the group consisting of S183A, S183T, S183V, S183Y, S183F, S183H, S183N, S183D, S183E, S183R, and S183K, and / or the V133 substitution is selected from the group consisting of V133E, V133S, V133L, V133W, V133K, V133R, and V133D. In some embodiments, the second antibody light chain comprises a third mutation and the second antibody heavy chain comprises a fourth mutation, and the third and fourth mutations promote selective association of the second antibody light chain with the second antibody heavy chain. In some embodiments, the third mutation comprises an amino acid substitution at V133 and / or the fourth mutation comprises an amino acid substitution at S183 (numbering according to the EU index). In some embodiments, the S183 substitution is selected from the group consisting of S183A, S183T, S183V, S183Y, S183F, S183H, S183N, S183D, S183E, S183R, and S183K, and / or the V133 substitution is selected from the group consisting of V133E, V133S, V133L, V133W, V133K, V133R, and V133D. In some embodiments, the amino acid substitution at S183 results in a positively charged residue (e.g., S183K) and the amino acid substitution at V133 results in a negatively charged residue (e.g., V133E). In some embodiments, the amino acid substitution at S183 results in a negatively charged residue (eg, S183E) and the amino acid substitution at V133 results in a positively charged residue (eg, V133K).In some embodiments, the first antibody light chain comprises a V133K mutation, the first antibody heavy chain comprises a S183E mutation, the second antibody light chain comprises a V133E mutation, and the second antibody heavy chain comprises a S183K mutation (numbering according to the EU index). In some embodiments, the first antibody heavy chain further comprises a T366S, L368A, and Y407V mutation, and the second antibody heavy chain further comprises a T366W mutation (numbering according to human IgG1 according to the EU index). In some embodiments, the second antibody light chain comprises a V133K mutation, the second antibody heavy chain comprises a S183E mutation, the first antibody light chain comprises a V133E mutation, and the first antibody heavy chain comprises a S183K mutation (numbering according to the EU index). In some embodiments, the second antibody heavy chain further comprises a T366S, L368A, and Y407V mutation and the first antibody heavy chain further comprises a T366W mutation (numbering based on human IgG1 according to the EU index).

[0020]

[0020] Another aspect of the disclosure relates to a method for identifying a combination of translation initiation sequences for expressing a multimeric polypeptide in a eukaryotic host cell, wherein the multimeric polypeptide comprises a first subunit comprising a first polypeptide chain and a second polypeptide chain, and a second subunit comprising a third polypeptide chain and a fourth polypeptide chain. In some embodiments, the method includes providing a library comprising a plurality of eukaryotic host cells, each of the plurality of eukaryotic host cells comprising a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding the first polypeptide chain, a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding the second polypeptide chain, a third polynucleotide comprising a third translation initiation sequence operably linked to a third open reading frame encoding the third polypeptide chain, and a fourth translation initiation sequence operably linked to a fourth open reading frame encoding the fourth polypeptide chain. and a fourth polynucleotide comprising a translation initiation sequence, wherein a plurality of combinations of the first, second, third, and fourth translation initiation sequences are represented in a plurality of eukaryotic host cells; culturing the library of eukaryotic host cells under conditions suitable for expression of the multimeric polypeptide by the plurality of eukaryotic host cells; measuring the amount of multimeric polypeptide expressed by a single eukaryotic host cell of the plurality of eukaryotic host cells or a clone of a single eukaryotic host cell of the plurality of eukaryotic host cells; and identifying the first, second, third, and fourth translation initiation sequences of one or more of the single eukaryotic host cells of the plurality of eukaryotic host cells or clones of a single eukaryotic host cell of the plurality of eukaryotic host cells that express the multimeric polypeptide. In some embodiments, the first, second, third, and fourth translation initiation sequences of one or more single eukaryotic host cells of a plurality of eukaryotic host cells or clones of a single eukaryotic host cell of a plurality of eukaryotic host cells expressing a multimeric polypeptide are identified based on increased production of multimeric polypeptides and / or reduced levels of mismatched by-products, e.g., compared to a reference or different combinations of four translation initiation sequences.In some embodiments, the first, second, third, and fourth translation initiation sequences in each host cell of the plurality of host cells all comprise the sequence (5' to 3') NNNNNATGNGA, where N is C, G, A, or T / U (SEQ ID NO:1).

[0021]

[0021] Another aspect of the present disclosure relates to a kit or article of manufacture comprising a polynucleotide for expressing a multimeric polypeptide comprising a first subunit comprising a first polypeptide chain and a second polypeptide chain, and a second subunit comprising a third polypeptide chain and a fourth polypeptide chain. In some embodiments, the kit or article of manufacture comprises a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding a first polypeptide chain, a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding a second polypeptide chain, a third polynucleotide comprising a third translation initiation sequence operably linked to a third open reading frame encoding a third polypeptide chain, and a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to a fourth open reading frame encoding a fourth polypeptide chain, wherein one or more of the first, second, third, and fourth translation initiation sequences are not operably linked to their respective open reading frames when each open reading frame is present in a naturally occurring host cell genome. In some embodiments, the multimeric polypeptide is a bispecific antibody, wherein the first and third polypeptide chains are antibody heavy chains, the second and fourth polypeptide chains are antibody light chains, the first subunit is a first half antibody that binds to a first antigen, and the second subunit is a second half antibody that binds to a second antigen.

[0022] In some embodiments, when each subunit is produced individually, one or both polypeptide chains of the first subunit are produced at a lower level than one or both polypeptide chains of the second subunit. In some embodiments, when all subunits are produced by the same host cell, one or both polypeptide chains of the first subunit are produced at a lower level than one or both polypeptide chains of the second subunit. In some embodiments, when each subunit is expressed individually in a eukaryotic host cell, the first subunit combines with more mismatched by-products than the second subunit. In some embodiments, when all subunits are expressed together in the same eukaryotic host cell, the first subunit combines with more mismatched by-products than the second subunit.

[0023] In some embodiments of any of the embodiments described herein, the first, second, third, and fourth translation initiation sequences all comprise the sequence (5' to 3') NNNNNATGNGA, where N is C, G, A, or T / U (SEQ ID NO:1). In some embodiments, one or both of the first translation initiation sequence and the second translation initiation sequence comprise a sequence selected from the group consisting of SEQ ID NOs:8-10. In some embodiments, one or both of the third translation initiation sequence and the fourth translation initiation sequence comprise the sequence ACCATGG (SEQ ID NO:3) or GAAGTATGA (SEQ ID NO:11). In some embodiments, the first translation initiation sequence comprises a sequence selected from the group consisting of SEQ ID NOs:8-10, the second translation initiation sequence comprises a sequence selected from the group consisting of SEQ ID NOs:8-10, the third translation initiation sequence comprises the sequence of SEQ ID NO:2, and the fourth translation initiation sequence comprises the sequence of SEQ ID NO:11. In some embodiments, the first translation initiation sequence comprises the sequence of SEQ ID NO:9, the second translation initiation sequence comprises the sequence of SEQ ID NO:9, the third translation initiation sequence comprises the sequence of SEQ ID NO:2, and the fourth translation initiation sequence comprises the sequence of SEQ ID NO:11. In some embodiments, each of the first, second, third, and fourth polynucleotides is operably linked to a promoter. In some embodiments, the first and second polynucleotides are operably linked to the same promoter, and the third and fourth polynucleotides are operably linked to the same promoter. In some embodiments, the first translation initiation sequence is weaker than the third translation initiation sequence. In some embodiments, the second translation initiation sequence is weaker than the fourth translation initiation sequence. In some embodiments, the first translation initiation sequence is weaker than the fourth translation initiation sequence. In some embodiments, the second translation initiation sequence is weaker than the third translation initiation sequence. In some embodiments, the first translation initiation sequence is the same as the second translation initiation sequence. In some embodiments, the third translation initiation sequence is the same as the fourth translation initiation sequence. In some embodiments of any of the embodiments described herein, the multimeric polypeptide specifically binds to one or more target antigens. In some embodiments, the multimeric polypeptide is a multispecific antigen-binding protein.

[0024]

[0024] Also provided herein is a polynucleotide comprising an open reading frame operably linked to a translation initiation sequence selected from the group consisting of SEQ ID NOs: 8-11. Provided herein is a polynucleotide comprising a first translation initiation sequence comprising a sequence selected from the group consisting of SEQ ID NOs: 8-10 operably linked to a first open reading frame, a second translation initiation sequence comprising a sequence selected from the group consisting of SEQ ID NOs: 8-10 operably linked to a second open reading frame, a third translation initiation sequence comprising the sequence of SEQ ID NO: 2 operably linked to a third open reading frame, and a fourth translation initiation sequence comprising the sequence of SEQ ID NO: 11 operably linked to a fourth open reading frame. Further provided herein is a set of polynucleotides. In some embodiments, the first translation initiation sequence comprises the sequence of SEQ ID NO:9, the second translation initiation sequence comprises the sequence of SEQ ID NO:9, the third translation initiation sequence comprises the sequence of SEQ ID NO:2, and the fourth translation initiation sequence comprises the sequence of SEQ ID NO:11. Further provided herein is a host cell comprising a polynucleotide or set of polynucleotides according to any of the embodiments described herein.

[0025] In some embodiments of any of the embodiments described herein, the first, second, third, and fourth polynucleotides are integrated into one or more chromosomes of the eukaryotic host cell. In some embodiments, the first, second, third, and fourth polynucleotides are integrated into the same chromosomal locus of the eukaryotic host cell. In some embodiments, the first, second, third, and fourth polynucleotides are part of one or more extrachromosomal polynucleotides in the eukaryotic host cell. In some embodiments, the eukaryotic host cell is a mammalian host cell. In some embodiments, the mammalian host cell is a Chinese Hamster Ovary (CHO) cell.

[0026]

[0026] It should be understood that one, some, or all of the features of the various embodiments described herein can be combined to form other embodiments of the present invention. These and other aspects of the present invention will be apparent to those skilled in the art. These and other embodiments of the present invention are further described by the following detailed description.

[0027] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the U.S. Patent and Trademark Office upon request and payment of the necessary fee. [Brief description of the drawings]

[0028] [Figure 1]

[0028] The names and nucleotide sequences of Kozak (Kz) sequence variants designed and analyzed by transient transfection are shown. The start codon is gray and underlined. The sequences shown correspond to SEQ ID NOs: 3-7 (top to bottom). [Diagram 2]

[0029] Normalized potency of Fc fusion proteins A and B with different Kozak sequence variants tested by transient transfection followed by homogeneous time-resolved fluorescence (HTRF). Data are shown as mean values ​​and error bars represent standard deviation (SD) of four independent transfections. Changes in the Kozak consensus sequence modulated the potency of the Fc fusion proteins. [Diagram 3]

[0030] We describe a process for designing a Kozak library, transforming E. coli cells with the library, sequencing to determine the nucleotide sequence of the variants, screening the variants by transient transfection, classifying and narrowing down the variants within a range, and selecting variants that cover the entire expression range. [Figure 4]

[0031] The wild type (WT) Kozak sequence is shown on the top line. On the bottom line, the designed Kozak library is shown. N is either A, T, G, or C. The start codon is grey and underlined. Positions -3 and +4 from the start codon are indicated. The sequences shown correspond to SEQ ID NOs: 23 and 1 (top to bottom), respectively. [Diagram 5]

[0032] Normalized antibody titers of 111 Kozak sequence variants transiently transfected into CHO cells are shown. Data are presented as mean values, and error bars represent standard deviation (SD) of two independent transfections. A nearly continuous range of translation was observed, with normalized titers ranging from 0.05 to 1.54 units (1.0 corresponds to the titer produced using the Wt Kozak sequence). [Figure 6]

[0033] Figure 1 shows the nucleotide distribution of 108 colonies of the Kozak library. The percentage of each nucleotide at each position is shown. The nucleotide distribution of the analyzed variants was random and diverse from position to position, indicating that the library screening was not skewed. [Figure 7]

[0034] Figure 1 shows the screening process by transient transfection in CHO cells performed to reduce the number of Kozak variants. The criteria for round-to-round selection of variants were reproducibility between transient transfections, stability between DNA preparations, and conservation of nucleotide diversity. [Figure 8A-8B]

[0035] Normalized titers and sequences of the final panel of Kozak sequence variants are shown. FIG. 8A shows normalized antibody titers of the indicated Kozak sequence variants transiently transfected into CHO cells. Data are shown as mean values, and error bars represent standard deviations of 11 independent transfections. The 11 Kozak sequence variants show an expression range of 0.2 to 1.3-fold that of Wt Kozak, allowing for titer adjustment with high precision. Asterisks indicate Kozak sequence variants that were selected as a representative panel of five variants, including Wt Kozak, because they covered the wide range of expression levels observed in transient transfections. FIG. 8B shows sequences of the final panel of Kozak sequence variants. The Wt consensus Kozak sequence is boxed. [Figure 9]

[0036] We show that by manipulating the light chain ratio (LC1:LC2), a high percentage of correctly assembled BsAbs can be obtained. For example, modifying the LC1:LC2 ratio from 1:2.5 to 1.5:1 increases correctly paired light chains by 40%. [Figure 10]

[0037] The translational strength of the listed Kozak variants relative to Wt Kozak (1x) is shown. The sequences of the listed Kozak variants are shown on the right. These Kozak sequence variants were selected as a representative panel of five variants, including Wt Kozak, because they covered the wide range of expression levels observed in transient transfections. The sequences correspond to SEQ ID NOs: 8, 9, 10, 2, and 11 (top to bottom). [Figure 11]

[0038] The Kozak variant combinations used to generate 25 integrative expression vectors for each group of Ab1 / Ab2 BsAbs are shown. A mixture of 50 plasmids was transfected into CHO cells. The diversity generated with this approach amounted to a potential 625 different chain ratio combinations. Only the combinations for Ab1 are shown, but the same combinations were used to generate the vectors for the Ab2 group. [Figure 12]

[0039] Transfection efficiency and transfection numbers of the Kozak mixed pools shown are shown. Transfection efficiency is monitored by performing cell surface staining analysis with allophycocyanin (APC)-labeled anti-hu IgG antibody and measuring antibody-APC expression by flow cytometry. Empty host was used to gate on Ab-expressing negative cells and GFP-positive cells. X-axis shows GFP expression and Y-axis shows APC-antibody expression. Similar transfection efficiency was observed for both Kozak mixed pools and Wt Kozak pools. [Figure 13A-13B]

[0040] Figure 1 shows the time course of stable pool recovery. Two pools per condition are shown. The addition of selection drug to establish stable pools is indicated by the arrow. The Y-axis shows the survival rate and the X-axis shows the number of days after addition of selection drug. [Figure 14A-14B]

[0041] Absolute titers of 704 individual clones of Kozak mixed clones or Wt Kozak clones selected from each transfection and condition and subjected to single cell cloning are shown. Titers were measured by HTRF and are shown in pairs based on transfection number as indicated. [Figure 15A-15B]

[0042] Figure 14 shows the viability of Kozak mixed clones and Wt-Kozak clones for 14-day shake flask fed-batch production. X-axis shows days and Y-axis shows viability. Transfection numbers are indicated. All clones investigated showed comparable viability throughout the production assay. Final viability on day 14 was approximately 80-95% in both conditions, except for one Kozak mixed clone, which resulted in a viability of 74%. [Figure 16A-16B]

[0043] Viable cell counts of 14-day shake flask fed-batch production of Kozak mixed clones and Wt-Kozak clones. X-axis indicates days and Y-axis indicates viable cells per milliliter / 106. Transfection numbers are shown. Exponential growth rates were observed up to day 7, followed by a plateau phase for all individual clones. [Figure 17]

[0044] Typical absolute titers of the top 11 Kozak mixed clones and Wt Kozak clones for each transfection are shown. Titers were measured on day 14 of shake flask fed-batch production. Individual Wt Kozak clones showed higher typical absolute antibody titers than the individual Kozak mixed clones in both transfections. Note that the titers represented correctly assembled bispecific antibodies as well as half antibodies and other unwanted by-products. [Figure 18]

[0045] Evaluate the quality of the top 11 Kozak mixed and Wt Kozak assembled antibodies for each transfection. Samples were analyzed by non-reducing capillary electrophoresis sodium dodecyl sulfate (CE-SDS) to distinguish and quantify the sum of the main peaks, which equated to full antibodies (full-ab) for the two antibody formats: correctly assembled and incorrectly assembled BsAb formats, and pre-peaks, which represent other species of different molecular weights. The number of transfections is shown. Within the top 22 clones, there were fewer Kozak mixed clones with a higher percentage of full-ab formation than the Wt Kozak clones. [Figure 19A-19B]

[0046] The quality and assembly efficiency of Kozak mixed clones with medium (represented by M) and low (represented by L) HCCF titers are shown. Figure 19A shows the typical absolute titers of three medium and four low Kozak mixed clones measured by shake flask fed-batch production assay on day 14. The production titers of these clones showed a similar behavior as previously observed in the primary screen. Figure 19B evaluates the quality of assembled antibodies of medium and low Kozak mixed clones. Samples were analyzed by non-reducing CE-SDS to distinguish and quantify the main peak equated to full antibody (full-ab) in the form of mispaired by-products and correctly assembled BsAb, and the sum of pre-peaks representing other species of different molecular weights. There was no direct correlation between the typical titers and the highest content of full antibody. For example, some Kozak mixed clones with more moderate titers compared to the top 11 clones showed a similar percentage of complete ab, around 50-60%. [Figure 20A-20B]

[0047] The quality and effective BsAb titers of the indicated Wt Kozak and Kozak mixed clones are shown. Figure 20A quantifies the species detected by mass spectrometry for 8 Kozak mixed clones and 8 Wt Kozak clones. The structure of each species is shown in the legend. Correctly assembled BsAb is highlighted in white. The best individual Kozak mixed clone showed about 40% of correctly assembled BsAb, which is more than twice as much as the best individual Wt Kozak clone (about 18%). Figure 20B shows the estimated titers of the bispecific antibodies and various mismatched by-product and half antibody species. The estimated titers were calculated as the percentage of each format multiplied by the general titer. The effective titers of the bispecifics are highlighted in white. Due to the fact that some Kozak mixed clones have higher assembly of BsAb than Wt Kozak clones, some Kozak mixed clones overcome the lack of general titers shown earlier in Figure 17. [Figures 21A-21C]

[0048] The Kozak sequence variant combinations and translation strengths of the indicated Kozak mixed clones are shown. Figure 21A shows the Kozak sequence variant combinations of the Kozak mixed clones. Highlighted in grey are the sequencing data that could not be obtained. The name and translation strength of each Kozak sequence variant are shown on the right. Figure 21B shows the Kozak sequence variant combinations of the best antibody producing Kozak mixed clones. The name and transfection number of the Kozak mixed clones are shown. Highlighted in grey are the sequencing data that could not be obtained. Figure 21C shows the Kozak sequence variant combinations of three medium Kozak mixed clones (designated M) and four low Kozak mixed clones (designated L) on the left. The name and translation strength of each Kozak sequence variant are shown on the right. Highlighted in grey are the sequencing data that could not be obtained. [Figure 22A]

[0049] Quantification of bispecific antibodies, mismatched by-products, and half-antibody species by mass spectrometry of Wt Kozak clone 50 and Kozak mixed clones 69 and 17M. Correctly assembled bispecific antibodies are highlighted in white. [Figure 22B]

[0050] The effective titers of bispecific antibodies, mismatched by-products, and half-antibody species, calculated as the percentage of each species multiplied by the general titer, are shown on the right. Correctly assembled bispecific antibodies are highlighted in white. [Figure 22C]

[0051] The production performance of clone 17M is shown above. The CE-SDS data of this clone indicates that the product quality is about 48% of full ab and about 52% of half ab. The mass spectrometry data showed that of the about 48% full ab, all of which represented correctly assembled bispecific antibodies. Meanwhile, almost all of the half ab species were knob 1 / 2 Ab2. This is consistent with the fact that the heavy and light chains of Ab2 were under the Wt Kozak sequence and the strongest Kozak sequence variant #228, respectively. The name and translation strength of each Kozak sequence variant are shown below. Clone 17M possessed both the heavy and light chains of Ab1 under a weak Kozak sequence variant. [Figure 22D]

[0052] Bispecific antibodies, mismatched by-products, and half-antibody formats are quantified by mass spectrometry for the medium and low Kozak mixed clones shown. The percentage of BsAb assemblies was less than 10%, except for Kozak mixed clone 17. The absolute abundance of each species correlated with the expression intensity of each chain. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0029] I. Definition

[0053] Before describing the present disclosure in detail, it is to be understood that this disclosure is not limited to particular compositions or biological systems, which can, of course, vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to be limiting.

[0030]

[0054] As used in this specification and the appended claims, the singular forms "a," "an," and "the" include plural referents unless the content clearly dictates otherwise. Thus, for example, reference to a "molecule" includes any combination of two or more such molecules.

[0031]

[0055] The term "about" as used herein refers to a normal range of error for the respective value, which would be readily understood by one of ordinary skill in the art. Reference herein to "about" a value or parameter includes (and describes) embodiments that are directed to that value or parameter itself.

[0032]

[0056] It is to be understood that aspects and embodiments of the present disclosure described herein include "comprising," "consisting," and "consisting essentially of" aspects and embodiments.

[0033]

[0057] "Polypeptide" or "protein" are used interchangeably herein to refer to polymers of amino acids of any length. The polymers may be linear or branched, may contain modified amino acids, and may be interrupted by non-amino acids. These terms also encompass amino acid polymers that are modified naturally or by intervention, such as disulfide bond formation, glycosylation, lipidation, acetylation, phosphorylation, or any other manipulation or modification, such as conjugation with a labeling moiety or a toxin. For example, polypeptides containing one or more analogs of an amino acid (including, for example, unnatural amino acids, etc.), and other modifications known in the art, are also included in this definition. The terms "polypeptide" and "protein" as used herein specifically encompass antibodies and antigen-binding proteins.

[0034]

[0058] As used herein, a "multimeric" polypeptide or protein may refer to any complex comprising more than one polypeptide or protein monomer (e.g., polypeptide chain). For example, a multimeric polypeptide or protein may refer to a complex comprising two or more subunits, each subunit comprising one, two, or more different polypeptide chains. The term "multimeric" polypeptide or protein specifically encompasses antibodies and antigen-binding proteins. In certain embodiments, a multimeric polypeptide is a bispecific antibody. In certain embodiments, a bispecific antibody comprises a first heavy chain, a first light chain, a second heavy chain, and a second light chain. In certain embodiments, a bispecific antibody is an IgG1, IgG2, or IgG4 isotype. In certain embodiments, a bispecific antibody is an IgG1 or IgG4 isotype. In certain embodiments, a bispecific antibody is an IgG1 isotype.

[0035]

[0059] The term "subunit" as used herein when used in reference to a component of a multimeric polypeptide or protein is intended to refer to any polypeptide containing more than one distinct polypeptide chain. In some embodiments, a subunit may comprise a macromolecular complex of two or more polypeptides bound together via one or more intermolecular bonds, including but not limited to one or more disulfide bonds. In certain embodiments, the subunit comprises an immunoglobulin heavy and light chain. In certain embodiments, the subunit comprises a half antibody. In certain embodiments, the immunoglobulin heavy and light chains or half antibody are of the IgG1, IgG2, or IgG4 isotype. In certain embodiments, the immunoglobulin heavy and light chains or half antibody are of the IgG1 or IgG4 isotype. In certain embodiments, the immunoglobulin heavy and light chains or half antibody are of the IgG1 isotype.

[0036]

[0060] As used herein, a "mismatched by-product" refers to the assembly of components of one or more polypeptide chains of a multimeric polypeptide in an undesired configuration. For example, if the multimeric polypeptide is a bispecific antibody comprising subunits light chain A / heavy chain A associated with subunits light chain B / heavy chain B, the mismatched by-product may result from mismatching during production of the heavy chain A dimer, the heavy chain B dimer, and / or mismatching of one or more of the light chain A / heavy chain A and light chain B / heavy chain B subunits, and may include any product other than a correctly assembled bispecific antibody. Specific examples of mismatched by-products are described with reference to and illustrated in Figures 20A, 20B, and 22A-22D. Other methods for assessing protein folding, such as Differential Scanning Fluorometry, MS, or methods for predicting production outcomes using in silico tools may also be used.

[0037]

[0061] As used herein, the term "translation initiation sequence" refers to a polynucleotide sequence that includes a start codon and adjacent bases of a polynucleotide that includes an open reading frame. According to the translation scanning model, the ribosome preinitiation complex binds to the 5' end of a polynucleotide and proceeds linearly in the 3' direction in search of a stop codon. The translation initiation sequence may be a "Kozak" sequence based on the work of Marilyn Kozak, who details a scanning model of translation initiation (e.g., Kozak M. Nucleic Acids Res. 1981;9(20):5233-52;Kozak M. Nucleic Acids Res. 1987;15(20):8125-48;Kozak M. EMBO J. 1997;16(9):2482-92;Kozak M. J Mol Biol. 1987;196(4):947-50;Hamilton R, Watanabe CK, de Boer HA. Nucleic Acids Res. 1987;15(8):3581-93;Cavener DR. Nucleic Acids Res. 1987;15(4):1353-61;Kozak M. Nucleic Acids Res. 1984;12(2):857-72; Gupta P, Rangan L, Ramesh TV, Gupta M. J Theor Biol. 2016;404:303-11; Grzegorski SJ, Chiari EF, Robbins A, Kish PE, Kahana A. PLoS One. 2014;9(9):e108475; and Kozak M. Cell. 1986;44(2):283-92), and variant Kozak sequences as described herein.

[0038] The terms "native" and "non-native" as used herein with respect to one or more genetic elements (e.g., encoding a polypeptide, a promoter, a translation unit, or a combination thereof) are intended to refer to the genomic context of the genetic elements in a host cell chromosome in which they occur in nature. For example, a polypeptide (e.g., a multimeric polypeptide) is "native" with respect to a host cell or a host cell chromosome when a polynucleotide encoding the polypeptide naturally occurs in the genome of the host cell, and is "non-native" when a polynucleotide encoding the polypeptide does not naturally occur in the genome of the host cell. A translation initiation sequence or an open reading frame is "native" with respect to a host cell or a host cell chromosome when the translation initiation sequence or the open reading frame naturally occurs in the genome of the host cell, and is "non-native" when the translation initiation sequence or the open reading frame does not naturally occur in the genome of the host cell. An operable combination of a translation initiation sequence and an open reading frame is "non-native" when the translation initiation sequence does not naturally occur in the genome of the host cell in the same operable linkage as the open reading frame, or vice versa. For example, a translation initiation sequence:open reading frame combination is "non-native" with respect to a host cell or host cell chromosome when one or both of the translation initiation sequence and the open reading frame do not naturally occur in the host cell genome, when a translation initiation sequence is present in the host cell genome operably linked to an open reading frame with which it is not operably associated in the naturally occurring host cell genome (even if the same translation initiation sequence is naturally present elsewhere in the host cell genome), or when an open reading frame is present in the host cell genome operably linked to a translation initiation sequence with which it is not operably associated in the naturally occurring host cell genome (even if the same open reading frame sequence is naturally present elsewhere in the host cell genome).

[0039]

[0063] The term "vector" as used herein is intended to refer to a nucleic acid molecule capable of transporting another nucleic acid to which it is linked. One type of vector is a "plasmid," which refers to a circular double stranded DNA loop to which additional DNA segments can be ligated. Another type of vector is a phage vector. Another type of vector is a viral vector, to which additional DNA segments can be ligated into the viral genome. Certain vectors are capable of autonomous replication in a host cell into which they are introduced (e.g., vectors with an origin of replication and episomal mammalian vectors). Other vectors (e.g., non-episomal mammalian vectors) can be integrated into the genome of a host cell upon introduction into the host cell, and thereby are replicated along with the host genome. In addition, certain vectors can induce the expression of genes to which they are operatively linked. Such vectors are referred to herein as "recombinant expression vectors" (or simply "recombinant vectors"). In general, expression vectors utilized in recombinant DNA techniques are often in the form of plasmids. As used herein, "plasmid" and "vector" can be used interchangeably, as the plasmid is the most commonly used form of vector.

[0040] "Operably linked" refers to the juxtaposition of two or more components, the components so described being in a relationship that allows them to function in their intended manner. Generally, "operably linked" DNA sequences are not necessarily contiguous, although, as necessary to join two protein coding regions or in the case of a secretory leader, they are contiguous and in reading frame. However, an operably linked promoter is generally located upstream of the coding sequence or translation unit, but not necessarily contiguous with it. An operably linked enhancer may be located upstream, within, or downstream of the coding sequence / translation unit, at a significant distance from the promoter. Linking is accomplished by recombinant methods known in the art, for example, using PCR methods, by annealing, or by ligation at convenient restriction sites. If no convenient restriction sites exist, synthetic oligonucleotide adaptors or linkers are conventionally used.

[0041]

[0065] "Promoter" refers to a polynucleotide sequence that controls transcription of a gene or sequence to which it is operably linked. A promoter contains signals for the binding of RNA polymerase and the initiation of transcription. The promoter used will be functional in a host cell of the cell type in which expression of the selected sequence is intended. Numerous promoters, including constitutive, inducible and repressible promoters, are known in the art (and identified in databases such as GenBank).

[0042]

[0066] The term "host cell" (or "recombinant host cell"), as used herein, is intended to refer to a cell that has been genetically modified or is capable of being genetically modified by the introduction of an exogenous or non-naturally occurring polynucleotide, such as a recombinant plasmid or vector. It should be understood that such terms are intended to refer not only to the particular subject cell, but also to the progeny of such a cell. Because certain modifications may occur in successive generations, either due to mutation or environmental influences, such progeny may not, in fact, be identical to the parent cell, but are still included within the scope of the term "host cell" as used herein.

[0043]

[0067] The term "antigen" herein is used in the broadest sense and encompasses various forms of both polypeptide and non-polypeptide antigens, including, but not limited to, small peptide antigens, full-length protein antigens, carbohydrate antigens, lipid antigens, and nucleic acid antigens.

[0044]

[0068] The term "antibody" herein is used in the broadest sense and encompasses a variety of antibody structures, including, but not limited to, monoclonal antibodies, polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), and antibody fragments, so long as they exhibit the desired antigen-binding activity. The term "immunoglobulin" (Ig) is used interchangeably with antibody herein.

[0045]

[0069] An "isolated" antibody is one that has been identified and separated and / or recovered from a component of its natural environment. Contaminant components of its natural environment are substances that would interfere with the antibody's experimental, diagnostic, or therapeutic uses, including enzymes, hormones, and other proteinaceous or nonproteinaceous solutes. In some embodiments, the antibody is purified (1) to greater than 95% by weight, and in some embodiments, greater than 99% by weight, as determined, for example, by the Lowry method; (2) to a degree sufficient to obtain at least 15 residues of N-terminal or internal amino acid sequence, for example, using a spinning cup sequenator; or (3) to homogeneity by SDS-PAGE under reducing or nonreducing conditions, for example, using Coomassie blue or silver staining. An isolated antibody includes the antibody in situ within recombinant cells, since at least one component of the antibody's natural environment will not be present. Ordinarily, however, an isolated antibody will be prepared by at least one purification step.

[0046]

[0070] The term "constant domain" refers to the portion of an immunoglobulin molecule that has a more conserved amino acid sequence compared to the other portion of the immunoglobulin, the variable domain, which contains the antigen binding site. The constant domain is the heavy chain C H 1. C H 2 and C H It contains three (collectively, CH) domains, as well as a light chain CHL (or CL) domain.

[0047]

[0071] The "variable region" or "variable domain" of an antibody refers to the amino-terminal domains of the heavy or light chain of the antibody. The variable domain of the heavy chain is referred to as "V H The variable domain of the light chain may be referred to as "V L " These domains are generally the most variable parts of the antibody and contain the antigen-binding sites.

[0048]

[0072] The term "variable" refers to the fact that the sequences of certain parts of the variable domains vary widely among antibodies and are used in the binding and specificity of each particular antibody to its particular antigen. However, the variability is not evenly distributed throughout the variable domains of antibodies. It is concentrated in three segments called hypervariable regions (HVRs) in both the light and heavy chain variable domains. The more highly conserved parts of the variable domains are called framework regions (FRs). Natural heavy and light chain variable domains each contain four FR regions that largely adopt a beta-sheet configuration connected by three HVRs that form loops that connect and, in some cases, form part of the beta-sheet structure. The HVRs within each chain are held in close proximity to each other by the FR regions and, together with the HVRs from the other chain, contribute to the formation of the antigen-binding site of antibodies (see Kabat et al., Sequences of Proteins of Immunological Interest, Fifth Edition, National Institute of Health, Bethesda, Md. (1991)). The constant domains are not involved directly in binding an antibody to an antigen, but exhibit various effector functions, such as involvement of the antibody in antibody-dependent cellular toxicity.

[0049]

[0073] The "light chains" of antibodies (immunoglobulins) from any mammalian species can be assigned to one of two clearly distinct types, called kappa ("κ") and lambda ("λ"), based on the amino acid sequences of their constant domains.

[0050]

[0074] The term IgG "isotype" or "subclass" as used herein means any of the subclasses of immunoglobulins defined by the chemical and antigenic properties of their constant regions.

[0051]

[0075] Depending on the amino acid sequence of the constant domain of their heavy chains, antibodies (immunoglobulins) can be assigned to different classes. There are five main classes of immunoglobulins: IgA, IgD, IgE, IgG, and IgM, some of which can be further divided into subclasses (isotypes), e.g., IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2. The heavy chain constant domains corresponding to the different classes of immunoglobulins are called α, γ, ε, γ, and μ, respectively. The subunit structures and three-dimensional configurations of the different classes of immunoglobulins are well known and are generally described, for example, in Abbas et al., Cellular and Mol. Immunology, 4th Edition (WB Saunders, Co., 2000). An antibody may be part of a larger fusion molecule formed by covalent or non-covalent association of the antibody with one or more other proteins or peptides.

[0052]

[0076] The terms "full length antibody," "intact antibody," and "whole antibody" are used interchangeably herein to refer to an antibody in its substantially intact form, rather than an antibody fragment as described below. These terms specifically refer to an antibody having a heavy chain that includes an Fc region.

[0053]

[0077] An "antibody fragment" includes a portion of an intact antibody, preferably including its antigen-binding region. In some embodiments, the antibody fragments described herein are antigen-binding fragments. Examples of antibody fragments include Fab, Fab', F(ab')2, and Fv fragments, diabodies, linear antibodies, single-chain antibody molecules, and multispecific antibodies formed from antibody fragments.

[0054]

[0078] Papain digestion of antibodies produces two identical antigen-binding fragments called "Fab" fragments, each with a single antigen-binding site, and a residual "Fc" fragment, a name reflecting its ability to crystallize readily. Pepsin treatment yields an F(ab')2 fragment that has two antigen-binding sites and is still capable of cross-linking antigen.

[0055]

[0079] "Fv" is the minimum antibody fragment that contains a complete antigen-binding site. In one embodiment, a two-chain Fv species consists of a dimer of one heavy-chain variable domain and one light-chain variable domain in tight, non-covalent association. In a single-chain Fv (scFv) species, one heavy-chain variable domain and one light-chain variable domain can be covalently linked by a flexible peptide linker such that the light and heavy chains can associate in a "dimeric" structure similar to that in the two-chain Fv species. It is in this configuration that the three HVRs of each variable domain interact to define an antigen-binding site on the surface of the VH-VL dimer. Collectively, the six HVRs confer antigen-binding specificity to the antibody. However, even a single variable domain (or half of an Fv containing only three antigen-specific HVRs) has the ability to recognize and bind to an antigen, albeit with a lower affinity than the entire binding site.

[0056]

[0080] Fab fragments contain heavy and light chain variable domains, and also contain a light chain constant domain and the first heavy chain constant domain (CH1). Fab' fragments differ from Fab fragments in that they have a few additional residues at the carboxy terminus of the heavy chain CH1 domain including one or more cysteines from the antibody hinge region. Fab'-SH is the designation herein for Fab' in which the cysteine ​​residue(s) of the constant domains bear a free thiol group. F(ab')2 antibody fragments were originally produced as pairs of Fab' fragments that have hinge cysteines between them. Other chemical couplings of antibody fragments are also known.

[0057]

[0081] "Single-chain Fv" or "scFv" antibody fragments comprise the VH and VL domains of an antibody, wherein these domains are present in a single polypeptide chain. Generally, the scFv polypeptide further comprises a polypeptide linker between the VH and VL domains, which enables the scFv to form the desired structure for antigen binding. For a review on scFvs, see, e.g., Pluckthun, The Pharmacology of Monoclonal Antibodies, Vol. 113, Eds. Rosenburg and Moore, (Springer-Verlag, New York, 1994), pp. 269-315.

[0058]

[0082] The term "diabody" refers to an antibody fragment having two antigen-binding sites, which fragments comprise a heavy chain variable domain (VH) connected to a light chain variable domain (VL) in the same polypeptide chain (VH-VL). By using a linker that is too short to allow pairing between the two domains on the same chain, these domains are forced to pair with complementary domains on another chain, creating two antigen-binding sites. Diabodies can be bivalent or bispecific. Diabodies are more fully described, for example, in EP 404,097, WO 1993 / 01161, Hudson et al., Nat. Med. 9:129-134 (2003), and Hollinger et al., Proc. Natl. Acad. Sci. USA, 90:6444-6448 (1993). Triabodies and tetrabodies are also described in Hudson et al., Nat. Med. 9:129-134 (2003).

[0059]

[0083] The term "monoclonal antibody" as used herein refers to an antibody obtained from a population of substantially homogenous antibodies, e.g., the individual antibodies that make up the population are identical except for possible mutations that may be present in minor amounts, e.g., naturally occurring mutations. Thus, the modifier "monoclonal" indicates the character of the antibody as not being a mixture of distinct antibodies. In certain embodiments, such monoclonal antibodies typically include antibodies that include a polypeptide sequence that binds to a target, the target-binding polypeptide sequence being obtained by a process that includes the selection of a single target-binding polypeptide sequence from a plurality of polypeptide sequences. For example, the selection process can be the selection of a unique clone from a plurality of clones, such as a pool of hybridoma clones, phage clones, or recombinant DNA clones. It is understood that the selected target-binding sequence may be further modified, e.g., to improve affinity to the target, humanize the target-binding sequence, improve its production in cell culture, reduce its immunogenicity in vivo, create a multispecific antibody, etc., and that an antibody that includes a modified target-binding sequence is also a monoclonal antibody of the present disclosure. In contrast to polyclonal antibody preparations, which typically include different antibodies directed against different determinants (epitopes), each monoclonal antibody of a monoclonal antibody preparation is directed against a single determinant on an antigen. In addition to their specificity, monoclonal antibody preparations are advantageous in that they are typically uncontaminated by other immunoglobulins.

[0060]

[0084] The modifier "monoclonal" indicates the character of the antibody as being obtained from a substantially homogeneous population of antibodies, and is not to be construed as requiring production of the antibody by any particular method. For example, the monoclonal antibodies used in accordance with the present disclosure can be produced by, for example, expression in prokaryotic host cells, hybridoma methods (e.g., Kohler and Milstein, Nature, 256:495-97(1975); Hongo et al., Hybridoma, 14(3):253-260(1995), Harlow et al., Antibodies: A Laboratory Manual, (Cold Spring Harbor Laboratory Press, 2nd ed. 1988); Hammerling et al., in: Monoclonal Antibodies and T-Cell Hybridomas 563-681(Elsevier, NY, 1981)), recombinant DNA methods (see, e.g., U.S. Pat. No. 4,816,567), phage display techniques (e.g., Clackson et al., Nature, 352:624-628(1991); Marks et al., J. Mol. Biol. 222:581-597(1992);Sidhu et al., J. Mol. Biol. 338(2):299-310(2004);Lee et al., J. Mol. Biol. 340(5):1073-1093(2004);Fellouse, Proc. Natl. Acad. Sci. USA 101(34):12467-12472(2004); and Lee et al., J. Immunol. Methods 284(1-2):119-132 (2004)), and techniques for producing human or human-like antibodies in animals that have some or all of the human immunoglobulin loci or genes that encode human immunoglobulin sequences (see, e.g., WO 1998 / 24893; WO 1996 / 34096; WO 1996 / 33735; WO 1991 / 10741; Jakobovits et al., Proc. Natl. Acad. Sci.USA 90:2551(1993);Jakobovits et al., Nature 362:255-258(1993);Bruggemann et al., Year in Immunol. 7:33 (1993); US Patent No. 5,545,807; US Patent No. 5,545,806; US Patent No. 5,569,825; US Patent No. 5,625,126; 10:779-783(1992);Lonberg et al., Nature 368:856-859(1994);Morrison, Nature 368:812-813(1994);Fishwild et al., Nature Biotechnol. 14:845-851 (1996); Neuberger, Nature Biotechnol. 14:826 (1996); and Lonberg and Huszar, Intern. Rev. Immunol. 13:65-93 (1995)).

[0061] As used herein, the term "hypervariable region", "HVR" or "HV" refers to the region of an antibody variable domain that is sequence hypervariable and / or forms structurally defined loops. Generally, an antibody contains six HVRs, three in the VH (H1, H2, H3) and three in the VL (L1, L2, L3). In natural antibodies, H3 and L3 show the highest diversity among the six HVRs, and H3 in particular is thought to play a unique role in conferring fine specificity to antibodies. See, for example, Xu et al., Immunity 13:37-45 (2000); Johnson and Wu, in Methods in Molecular Biology 248:1-25 (Lo, ed., Human Press, Totowa, NJ, 2003). In fact, naturally occurring camelid antibodies, consisting only of heavy chains, are functional and stable in the absence of light chains. See, e.g., Hamers-Casterman et al., Nature 363:446-448 (1993); Sheriff et al., Nature Struct. Biol. 3:733-736 (1996).

[0062]

[0086] Several HVR delineations are used and encompassed herein. The Kabat complementarity determining regions (CDRs) are based on sequence variability and are the most commonly used (Kabat et al., Sequences of Proteins of Immunological Interest, 5th ed., Public Health Service, National Institutes of Health, Bethesda, Md. (1991)). Instead, Chothia refers to the location of structural loops (Chothia and Lesk J. Mol. Biol. 196:901-917 (1987)). The AbM HVRs represent a compromise between the Kabat HVRs and the Chothia structural loops and are used by Oxford Molecular's AbM antibody modeling software. The "contact" HVRs are based on an analysis of available complex crystal structures. Residues from each of these HVRs are shown below.

[0063] TIFF0007658905000001.tif88170

[0064]

[0087] HVRs may include the following "extended HVRs": in the VL, 24-36 or 24-34 (L1), 46-56 or 50-56 (L2), and 89-97 or 89-96 (L3), and in the VH, 26-35 (H1), 50-65 or 49-65 (H2), and 93-102, 94-102, or 95-102 (H3). The variable domain residues are numbered according to Kabat et al. (see above) for each of these definitions.

[0065]

[0088] "Framework" or "FR" residues are those variable domain residues other than HVR residues as herein defined.

[0066]

[0089] The terms "variable domain residue numbering as in Kabat" or "amino acid position numbering as in Kabat", and variations thereof, refer to the numbering system used for the heavy or light chain variable domains of the compilation of antibodies in Kabat et al. (see above). Using this numbering system, the actual linear amino acid sequence may contain fewer or additional amino acids corresponding to a shortening of, or insertion into, the FR or HVR of the variable domain. For example, a heavy chain variable domain may contain a single amino acid insertion after residue 52 of H2 (residue 52a according to Kabat) and may contain inserted residues after heavy chain FR residue 82 (e.g., residues 82a, 82b, and 82c, etc. according to Kabat). The Kabat numbering of residues may be determined for a given antibody by alignment of the antibody's sequence with the "standard" Kabat numbered sequence at the regions of homology.

[0067]

[0090] The Kabat numbering system is commonly used when referring to residues within the variable domain (approximately residues 1-107 in the light chain and residues 1-113 in the heavy chain) (e.g., Kabat et al., Sequences of Immunological Interest. 5th Ed. Public Health Service, National Institutes of Health, Bethesda, Md. (1991)). The "EU numbering system" or "EU index" is commonly used when referring to residues in an immunoglobulin heavy chain constant region (e.g., the EU index reported in Kabat et al., supra).

[0068]

[0091] The term "linear antibody" refers to the antibodies described in Zapata et al. (1995 Protein Eng, 8(10):1057-1062). Briefly, these antibodies comprise a pair of tandem Fd segments (VH-CH1-VH-CH1) which, together with complementary light chain polypeptides, form a pair of antigen-binding regions. Linear antibodies can be bispecific or monospecific.

[0069] II. Methods for Producing and Screening Multimeric Polypeptides

[0092] Provided herein are methods for producing multimeric polypeptides in eukaryotic host cells (eg, mammalian host cells). In some embodiments, the method includes providing a eukaryotic host cell comprising a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding a first polypeptide chain of a multimeric polypeptide, a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding a second polypeptide chain of the multimeric polypeptide, a third polynucleotide comprising a third translation initiation sequence operably linked to a third open reading frame encoding a third polypeptide chain of the multimeric polypeptide, and a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to a fourth open reading frame encoding a fourth polypeptide chain of the multimeric polypeptide, culturing the eukaryotic host cell under conditions suitable for expression of the first, second, third, and fourth polypeptide chains, wherein the first, second, third, and fourth polypeptide chains form a multimeric polypeptide, and recovering the multimeric polypeptide produced by the eukaryotic host cell.

[0070]

[0093] In some embodiments, the first subunit is produced at a lower level than the second subunit when each polypeptide chain is produced individually in a eukaryotic host cell, or when both polypeptide chains are produced together by the same eukaryotic host cell, one or both of the first translation initiation sequence and the second translation initiation sequence are weaker than one or both of the third translation initiation sequence and the fourth translation initiation sequence. In some embodiments, each subunit comprises a half antibody comprising an antibody heavy chain and an antibody light chain, and the multimeric polypeptide is a bispecific antibody.

[0071]

[0094] The production methods described herein are based, at least in part, on the demonstration herein that adjustment of translation strength using different translation initiation sequences can improve multimeric polypeptide (e.g., bispecific antibody) assembly and reduce product-related impurities such as mismatched by-products. Surprisingly, it has been found that reducing the translation of selected polypeptide chains can effectively increase multimeric polypeptide assembly relative to the translation levels of other polypeptide chains of the multimeric polypeptide and improve overall product quality by limiting the accumulation of mismatched by-products. Without wishing to be bound by theory, it is believed that the production of some polypeptide chains under the strongest translation initiation sequence actually promoted the production of more mismatched by-products containing that chain instead of more correctly assembled multimeric polypeptides. Thus, downregulation of the translation of one or more chains of a multimeric polypeptide such as a bispecific antibody, which can slow down the assembly of a multimeric polypeptide such as a bispecific antibody, can actually have a positive impact on the assembly efficiency and the reduction of undesired mismatched by-products. Thus, the methods described herein not only allow for high levels of multimeric polypeptide expression, but also reduce the production of impurities such as mismatched by-products that impose burdens such as the need for costly purification methods for their removal.

[0072]

[0095] In some embodiments, when each subunit is produced individually in a eukaryotic host cell or when all subunits are produced together in the same eukaryotic host cell, one or both polypeptide chains of the first subunit are translated at a lower level than one or both polypeptide chains of the second subunit. In some embodiments, when each subunit is produced individually in a eukaryotic host cell or when all subunits are produced together in the same eukaryotic host cell, the first subunit combines with more mismatched by-products than the second subunit. In some embodiments, the use of a weak translation initiation sequence operably linked to one or more (e.g., one or both) of the polypeptide chains of the first subunit (e.g., a subunit that is a weaker or more difficult expressor in the host cell) results in the production of higher yields of multimeric polypeptides and / or fewer mismatched by-products, compared to a translation initiation sequence associated with one or more (e.g., one or both) of the polypeptide chains of the second subunit. In some embodiments, when each subunit is expressed individually in a eukaryotic host cell, one or both polypeptide chains of the first subunit are translated at a slower rate than one or both polypeptide chains of the second subunit. In some embodiments, when each subunit is expressed individually in a eukaryotic host cell, one or both polypeptide chains of the first subunit are folded more slowly and / or less efficiently than one or both polypeptide chains of the second subunit. In some embodiments, when each subunit is expressed individually in a eukaryotic host cell, the first subunit is assembled at a slower rate than the second subunit. Assays for measuring protein translation rates are known in the art and include, but are not limited to, 35S methionine labeling and ribosome profiling (see, e.g., Ingolia, N. (2016) Cell 165:22-33).

[0073]

[0096] As described herein, in some embodiments, the multimeric polypeptide of the present disclosure comprises two subunits. In some embodiments, the subunit of the present disclosure comprises two or more polypeptide chains. In some embodiments, the subunit of the present disclosure comprises two polypeptide chains. In some embodiments, each polynucleotide encoding a polypeptide chain of the multimeric polypeptide comprises a translation initiation sequence operably linked to an open reading frame. In some embodiments, the multimeric polypeptide of the present disclosure comprises two subunits, each subunit comprising two polypeptide chains, the first subunit of the multimeric polypeptide is expressed at a lower level than the second subunit when each subunit is expressed individually in a eukaryotic host cell or when all subunits are expressed together in the same eukaryotic host cell, and one or both of the first translation initiation sequence (operably linked to the first open reading frame) and the second translation initiation sequence (operably linked to the second open reading frame) are weaker than one or both of the third translation initiation sequence (operably linked to the third open reading frame) and the fourth translation initiation sequence (operably linked to the fourth open reading frame). In some embodiments, the multimeric polypeptide is a bispecific antibody.

[0074]

[0097] In some embodiments, a subunit or half antibody of the disclosure is expressed at a lower level than another subunit or half antibody when each subunit or half antibody is expressed individually in a eukaryotic host cell. In some embodiments, a subunit or half antibody of the disclosure is expressed at a level that is at least 5%, at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, 5% to 30%, 5% to 50%, 5% to 75%, 10% to 30%, 10% to 50%, 10% to 75%, 25% to 50%, 25% to 75%, 25% to 100%, 50% to 75%, 50% to 100%, or 75% to 100% lower than another subunit or half antibody, e.g., when each subunit or half antibody is expressed individually in a eukaryotic host cell. In some embodiments, a subunit or half antibody of the disclosure is expressed at a level that is at least 1.3-fold, at least 1.5-fold, at least 2-fold, at least 2.5-fold, at least 3-fold, at least 3.5-fold, at least 4-fold, at least 4.5-fold, at least 5-fold, at least 6-fold, at least 7-fold, at least 8-fold, at least 9-fold, at least 10-fold, 1.3-fold to 3-fold, 1.5-fold to 3-fold, 2-fold to 10-fold, 2-fold to 5-fold, 3-fold to 5-fold, 3-fold to 10-fold, 5-fold to 10-fold, or 7-fold to 10-fold lower than another subunit or half antibody, e.g., when each subunit or half antibody is expressed individually in a eukaryotic host cell. In some embodiments, a subunit or half antibody of the disclosure is expressed at a level that is 0.2-fold to 0.8-fold, less than 0.8-fold, less than 0.5-fold, or less than 0.3-fold lower than another subunit or half antibody, e.g., when each subunit or half antibody is expressed individually in a eukaryotic host cell.

[0075]

[0098] In some embodiments, the method comprises providing a eukaryotic host cell comprising a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding a first antibody heavy chain of the bispecific antibody, a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding a first antibody light chain of the bispecific antibody, a third polynucleotide comprising a third translation initiation sequence operably linked to a third open reading frame encoding a second antibody heavy chain of the bispecific antibody, and a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to a fourth open reading frame encoding a second antibody light chain of the bispecific antibody, culturing the eukaryotic host cell under conditions suitable for expression of the first, second, third, and fourth polypeptide chains, wherein the first, second, third, and fourth antibody heavy chains form the bispecific antibody, culturing the eukaryotic host cell, and recovering the bispecific antibody from the eukaryotic host cell. In some embodiments, a bispecific antibody specifically binds to two antigens (e.g., a first antibody heavy and light chains form an antigen binding domain that binds a first antigen, and a second antibody heavy and light chains form an antigen binding domain that binds a second antigen).

[0076]

[0099] In some embodiments, a first translation initiation sequence is said to be weaker than another translation initiation sequence when the translation efficiency and / or expression of an open reading frame operably linked to the translation initiation sequence is lower compared to a reference or efficiency / expression of the same open reading frame operably linked to another translation initiation sequence. Methods suitable for comparing the strength of various translation initiation sequences in eukaryotic host cells are described and exemplified herein. For example, transient or stable transfectants can be cultured and the production of multimeric polypeptides can be measured. Fed-batch or perfusion cultures can be performed and the titer of the product in the cell medium or on the cell surface can be measured. For surface expression, cells can be stained with antibodies to detect the product and analyzed, for example, by flow cytometry. The quality and / or purity of the product can be evaluated, for example, using electrophoresis and / or mass spectrometry.

[0077]

[0100] In some embodiments, the translation initiation sequence is at least 5%, at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, 5% to 30%, 5% to 50%, 5% to 75%, 10% to 30%, 10% to 50%, 10% to 75%, 25% to 50%, 25% to 75%, 25% to 100%, 50% to 75%, 50% to 100%, or 75% to 100% weaker than an alternative translation initiation sequence. In some embodiments, the translation initiation sequence exhibits a translation efficiency of at least 5%, at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 100%, at least 110%, at least 120%, at least 130%, at least 140%, at least 150%, at least 160%, at least 170%, at least 180%, at least 190%, at least 210%, at least 220%, at least 230%, at least 240%, at least 250%, at least 260%, at least 270%, at least 280%, at least 290%, at least 300%, at least 310%, at least 320%, at least 330%, at least 340%, at least 350%, at least 360%, at least 370%, at least 380%, at least 390%, at least 400%, at least 410%, at least 420%, at least 430%, at least 440%, at least 450%, at least 460%, at least 470%, at least 480%, at least 490%, at least 500%, at least 510%, at least 520%, at least 530%, at least 540%, at least 550%, at least 550%, at least 560%, at least 570%, at least 580%, at least 590%, at least 600%, at least 610%, at least 620%, at least 630%, at least 640%, at least 650%, at least 650%, at least 700%, at least 750 A translation initiation sequence is said to be weaker than an alternative translation initiation sequence when it is 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, 5% to 30%, 5% to 50%, 5% to 75%, 10% to 30%, 10% to 50%, 10% to 75%, 25% to 50%, 25% to 75%, 25% to 100%, 50% to 75%, 50% to 100%, or 75% to 100% weaker.In some embodiments, the translation initiation sequence reduces expression of the same open reading frame operably linked to the translation initiation sequence by at least 5%, at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 100%, at least 120%, at least 140%, at least 160%, at least 180%, at least 190%, at least 210%, at least 220%, at least 230%, at least 240%, at least 250%, at least 300%, at least 350%, at least 400%, at least 450%, at least 500%, at least 1000%, at least 1200%, at least 1400%, at least 1600%, at least 1800%, at least 1900%, at least 2100%, at least 2200%, at least 2300%, at least 2400%, at least 2500%, at least 3000%, at least 3500%, at least 4000%, at least 4500%, at least 5000%, at least 1 ... A translation initiation sequence is said to be weaker than an alternative translation initiation sequence when it is at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 100%, 5% to 30%, 5% to 50%, 5% to 75%, 10% to 30%, 10% to 50%, 10% to 75%, 25% to 50%, 25% to 75%, 25% to 100%, 50% to 75%, 50% to 100%, or 75% to 100% weaker.

[0078]

[0101] In some embodiments, a translation initiation sequence is at least 1.3-fold, at least 1.5-fold, at least 2-fold, at least 2.5-fold, at least 3-fold, at least 3.5-fold, at least 4-fold, at least 4.5-fold, at least 5-fold, at least 6-fold, at least 7-fold, at least 8-fold, at least 9-fold, at least 10-fold, 1.3-fold to 3-fold, 1.5-fold to 3-fold, 2-fold to 10-fold, 2-fold to 5-fold, 3-fold to 5-fold, 3-fold to 10-fold, 5-fold to 10-fold, or 7-fold to 10-fold weaker than another translation initiation sequence. In some embodiments, a translation initiation sequence is said to be weaker than another translation initiation sequence when the translation efficiency of an open reading frame operably linked to the translation initiation sequence is at least 1.3-fold, at least 1.5-fold, at least 2-fold, at least 2.5-fold, at least 3-fold, at least 3.5-fold, at least 4-fold, at least 4.5-fold, at least 5-fold, at least 6-fold, at least 7-fold, at least 8-fold, at least 9-fold, at least 10-fold, 1.3-fold to 3-fold, 1.5-fold to 3-fold, 2-fold to 10-fold, 2-fold to 5-fold, 3-fold to 5-fold, 3-fold to 10-fold, 5-fold to 10-fold, or 7-fold to 10-fold lower compared to a reference or efficiency of the same open reading frame operably linked to another translation initiation sequence (e.g., when measured as described above). In some embodiments, a translation initiation sequence is said to be weaker than another translation initiation sequence when expression of an open reading frame operably linked to the translation initiation sequence is at least 1.3-fold, at least 1.5-fold, at least 2-fold, at least 2.5-fold, at least 3-fold, at least 3.5-fold, at least 4-fold, at least 4.5-fold, at least 5-fold, at least 6-fold, at least 7-fold, at least 8-fold, at least 9-fold, at least 10-fold, 1.3-fold to 3-fold, 1.5-fold to 3-fold, 2-fold to 10-fold, 2-fold to 5-fold, 3-fold to 5-fold, 3-fold to 10-fold, 5-fold to 10-fold, or 7-fold to 10-fold lower compared to a reference or expression level of the same open reading frame operably linked to another translation initiation sequence (e.g., when measured as described above).In some embodiments, a translation initiation sequence is said to be weaker than another translation initiation sequence when it reduces expression of an open reading frame operably linked to the translation initiation sequence by 0.2-fold to 0.8-fold, less than 0.8-fold, less than 0.5-fold, or less than 0.3-fold compared to a reference or expression level of the same open reading frame operably linked to another translation initiation sequence (e.g., when measured as described above).

[0079]

[0102] In some embodiments, the first translation initiation sequence is weaker than the third translation initiation sequence. In some embodiments, the second translation initiation sequence is weaker than the fourth translation initiation sequence. In some embodiments, the first translation initiation sequence is weaker than the fourth translation initiation sequence. In some embodiments, the second translation initiation sequence is weaker than the third translation initiation sequence. In some embodiments, the first translation initiation sequence is the same as the second translation initiation sequence. In some embodiments, the first translation initiation sequence is different from the second translation initiation sequence. In some embodiments, the third translation initiation sequence is the same as the fourth translation initiation sequence. In some embodiments, the third translation initiation sequence is different from the fourth translation initiation sequence.

[0080]

[0103] In some embodiments, the methods of the disclosure provide for higher production of multimeric polypeptides in eukaryotic host cells, e.g., compared to production of multimeric polypeptides using existing methods. In some embodiments, the methods of the disclosure provide for at least 5%, at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, or at least 100% higher production of multimeric polypeptides (e.g., higher levels of correctly assembled multimeric polypeptides, such as bispecific antibodies) in eukaryotic host cells, e.g., compared to production of multimeric polypeptides using existing methods. For example, expression can be compared to expression in which one or more, two or more, three or more, or four of the polynucleotides encoding the polypeptide chains of a multimeric polypeptide contain an open reading frame operably linked to a native or unmodified translation initiation sequence, or expression in which each of the polynucleotides encoding the polypeptide chains of a multimeric polypeptide contain the same translation initiation sequence.

[0081]

[0104] In some embodiments, the method of the present disclosure reduces the amount of mismatched by-products of multimeric polypeptides in eukaryotic host cells, for example, compared to the production of multimeric polypeptides using existing methods. In some embodiments, the method of the present disclosure reduces the amount of mismatched by-products of multimeric polypeptides in eukaryotic host cells, for example, compared to the production of multimeric polypeptides using existing methods, for example, compared to the production of multimeric polypeptides using existing methods. The method of the present disclosure is advantageous in performing down-regulation purification in production, because there are fewer impurities (e.g., mismatched by-products) in the cell culture supernatant. In some embodiments, the production can be compared to a production in which one or more, two or more, three or more, or four of the polynucleotides encoding the polypeptide chains of the multimeric polypeptides comprise an open reading frame operably linked to a native or unmodified translation initiation sequence, or in which each of the polynucleotides encoding the polypeptide chains of the multimeric polypeptides comprises the same translation initiation sequence. In some embodiments, the amount of mismatched by-products refers to the amount of one or more specific mismatched by-products. In some embodiments, the amount of mismatched by-products refers to the total amount of all mismatched by-products.

[0082]

[0105] In some embodiments, the methods of the present disclosure can be used to express the four subunits of the multimeric polypeptide of the present disclosure in a chain ratio of 1:1:1:1. However, the methods described herein can also be used to adjust the expression of the four subunits of the multimeric polypeptide of the present disclosure in other desired chain ratios, e.g., to improve the expression and / or assembly of a correctly paired multimeric polypeptide. For example, it is well documented that antibody light chains can function as chaperones for antibody heavy chains (see, e.g., Lee, Y. K, et al. (1999) Mol. Biol. Cell 10:2209-2219), and high expression of light chains can be advantageous during manufacturing. Thus, in some embodiments, the methods of the present disclosure are used to increase the expression of one or both light chains of a bispecific antibody relative to the expression of the heavy chain. Furthermore, the data provided herein demonstrates that in some cases, the ratio of light chain 1:light chain 2 can be manipulated to increase the yield of correctly assembled bispecific antibodies. Thus, in some embodiments, the methods of the disclosure are used to increase expression of one light chain of a bispecific antibody relative to the expression of the other light chain. In some embodiments, depending on the optimal method for producing a bispecific antibody, which may be determined herein, the methods of the disclosure may be used to express the four polypeptide chains of a bispecific antibody in a 1:1:1:1 chain ratio or another chain ratio.

[0083]

[0106] In some embodiments, one or more of the first, second, third, and fourth translation initiation sequences comprise the sequence (5' to 3') NNNNNATGNGA, where N is C, G, A, or T / U (SEQ ID NO:1). In some embodiments, all of the first, second, third, and fourth translation initiation sequences comprise the sequence (5' to 3') NNNNNATGNGA, where N is C, G, A, or T / U (SEQ ID NO:1). In some embodiments, one or both translation initiation sequences operably linked to an open reading frame encoding a polypeptide chain of a first subunit (e.g., a lesser expressed subunit of a multimeric polypeptide) comprise a sequence selected from the group consisting of SEQ ID NOs:8-10. In some embodiments, one or both translation initiation sequences operably linked to an open reading frame encoding a polypeptide chain of a second subunit (e.g., a more strongly expressed subunit of a multimeric polypeptide) comprise a sequence selected from the group consisting of SEQ ID NOs:2, 3, and 11.

[0084]

[0107] Exemplary translation initiation sequences are shown in Table 2. The relative strengths of exemplary translation initiation sequences are shown in FIG.

[0085] TIFF0007658905000002.tif95170

[0086]

[0108] In some embodiments, the translation efficiency (i.e., strength) of a translation initiation sequence is compared to a reference. In some embodiments, the reference is a translation initiation sequence that comprises a wild-type Kozak sequence in the host cell. In some embodiments, the reference sequence comprises a polynucleotide sequence of SEQ ID NO: 2 or 3. Exemplary assays suitable for determining the relative strength of a translation initiation sequence are described and exemplified herein (see, e.g., Examples 1-3). In some embodiments, the strength of a translation initiation sequence is assessed by expressing in a host cell (e.g., after transient transfection) a polynucleotide that comprises a translation initiation sequence operably linked to an open reading frame (e.g., encoding a polypeptide, subunit, or multimeric polypeptide of the disclosure, such as an antibody or half-antibody) and measuring the production of a polypeptide product. In some embodiments, the production level of a polypeptide product (e.g., product titer) is compared to a reference, such as the production level of the same product in the same host cell type where the open reading frame is operably linked to a reference translation initiation sequence, such as a wild-type Kozak sequence active in the host cell. The relative strength of exemplary translation initiation sequences is shown in FIG. 10. For example, transient or stable transfectants can be cultured and the production of multimeric polypeptides can be measured. Fed-batch or perfusion cultures can be performed and the titer of the product in the cell medium or on the cell surface can be measured. For surface expression, cells can be stained with antibodies to detect the product and analyzed, for example, by flow cytometry. The quality and / or purity of the product can be assessed, for example, using electrophoresis and / or mass spectrometry.

[0087]

[0109] In some embodiments, the open reading frame of the first polypeptide chain of the first subunit of the multimeric protein (e.g., the lesser expressed subunit) is operably linked to a translation initiation sequence comprising SEQ ID NO: 9. In some embodiments, the open reading frame of the second polypeptide chain of the first subunit of the multimeric protein (e.g., the lesser expressed subunit) is operably linked to a translation initiation sequence comprising SEQ ID NO: 9. In some embodiments, the open reading frame of the first polypeptide chain of the second subunit of the multimeric protein (e.g., the more strongly expressed subunit) is operably linked to a translation initiation sequence comprising SEQ ID NO: 2 or 11. In some embodiments, the open reading frame of the second polypeptide chain of the second subunit of the multimeric protein (e.g., the more strongly expressed subunit) is operably linked to a translation initiation sequence comprising SEQ ID NO: 2 or 11. In some embodiments, the open reading frame of the first polypeptide chain of the first subunit (e.g., the weaker expressed subunit) is operably linked to a translation initiation sequence comprising SEQ ID NO:9, the open reading frame of the second polypeptide chain of the first subunit (e.g., the weaker expressed subunit) is operably linked to a translation initiation sequence comprising SEQ ID NO:9, the open reading frame of the first polypeptide chain of the second subunit (e.g., the stronger expressed subunit) is operably linked to a translation initiation sequence comprising SEQ ID NO:2 or 11, and the open reading frame of the second polypeptide chain of the second subunit (e.g., the stronger expressed subunit) is operably linked to a translation initiation sequence comprising SEQ ID NO:2 or 11. In some embodiments, the multimeric protein is a bispecific antibody. In some embodiments, the first and second subunits comprise half antibodies. In some embodiments, the two polypeptide chains of each subunit represent an antibody heavy chain and an antibody light chain, respectively.In some embodiments, an open reading frame encoding the heavy chain of a first half antibody is operably linked to a translation initiation sequence comprising SEQ ID NO:9, an open reading frame encoding the light chain of the first half antibody is operably linked to a translation initiation sequence comprising SEQ ID NO:9, an open reading frame encoding the heavy chain of a second half antibody is operably linked to a translation initiation sequence comprising SEQ ID NO:3, and an open reading frame encoding the light chain of the second half antibody is operably linked to a translation initiation sequence comprising SEQ ID NO:11.

[0088] promoter

[0110] In some embodiments, the translation initiation sequence and / or open reading frame of the disclosure is operably linked to a promoter. In some embodiments, each of the first, second, third, and fourth polynucleotides is operably linked to a promoter. A promoter is a non-translated regulatory sequence located upstream (5') of a cistron that regulates its expression. Prokaryotic promoters are typically divided into two classes: inducible and constitutive. An inducible promoter is a promoter that initiates increased levels of transcription of a cistron under its control in response to a change in culture conditions, such as the presence or absence of a nutrient or a change in temperature.

[0089] In some embodiments, the promoter is a constitutive promoter. In some embodiments, the promoter is an inducible promoter. A variety of promoters suitable for use in eukaryotic host cells are known in the art. In some embodiments, the polynucleotides encoding the first and second polypeptide chains of the subunits are operably linked to the same promoter. In some embodiments, the polynucleotides encoding the first and second polypeptide chains of the subunits are operably linked to different promoters.

[0090]

[0112] A large number of promoters recognized by a variety of potential host cells are well known. The selected promoter can be operably linked to, for example, the cistron DNA encoding the light or heavy chain by removing the promoter from the source DNA by restriction enzyme digestion and inserting the isolated promoter sequence into the vector of the invention. Both the native promoter sequence and many heterologous promoters can be used to directly amplify and / or express the target gene. In some embodiments, heterologous promoters are utilized because they generally result in greater transcription and higher yields of the expressed target gene compared to the native target polypeptide promoter.

[0091] Expression and cloning vectors usually contain a promoter that is recognized by the host organism and operably linked to the desired Fc-containing polypeptide(s) (e.g., antibody) nucleic acid. Eukaryotic promoter sequences are known. Virtually all eukaryotic genes have an AT-rich region located approximately 25-30 bases upstream from the site where transcription begins. Another sequence found 70-80 bases upstream from the start of transcription of many genes is a CNCAAT region, where N can be any nucleotide. At the 3' end of most eukaryotic genes is an AATAAA sequence, which can be a signal for addition of a poly A tail to the 3' end of the coding sequence. All of these sequences are suitably inserted into eukaryotic expression vectors.

[0092]

[0114] For production of Fc-containing polypeptide(s) (such as, for example, antibodies), transcription from the vector in a mammalian host cell can be controlled by a promoter derived from the genome of a virus such as, for example, polyoma virus, fowlpox virus, adenovirus (such as adenovirus 2), bovine papilloma virus, avian sarcoma virus, cytomegalovirus, retrovirus, hepatitis B virus, and simian virus 40 (SV40), from a heterologous mammalian promoter, such as the actin promoter or immunoglobulin promoter, or from a heat shock promoter, provided that such promoter is compatible with the host cell system. In some embodiments, the promoter is a CMV promoter.

[0093] The early and late promoters of the SV40 virus are conveniently obtained as an SV40 restriction fragment which also contains the SV40 viral origin of replication. The immediate early promoter of the human cytomegalovirus is conveniently obtained as a Hind 111E restriction fragment. A system for expressing DNA in mammalian hosts using bovine papilloma virus as a vector is disclosed in U.S. Pat. No. 4,419,446. A modification of this system is described in U.S. Pat. No. 4,601,978. See also Reyes et al., Nature 297:598-601 (1982), for expression of human β-interferon cDNA in mouse cells under the control of the thymidine kinase promoter from herpes simplex virus. Alternatively, the Rous sarcoma virus long terminal repeat can be used as a promoter.

[0094]

[0116] Transcription of DNA encoding an antigen-binding polypeptide(s), such as an antibody, by higher eukaryotes can be increased by inserting an enhancer sequence into the vector. Many enhancer sequences are now known from mammalian genes, such as globin, elastase, albumin, a-fetoprotein, and insulin genes. Enhancers from eukaryotic cell viruses may also be used. Examples include the SV40 enhancer on the late side of the replication origin (bp 100-270), the cytomegalovirus early promoter enhancer, the polyoma enhancer on the late side of the replication origin, and adenovirus enhancers. See also Yaniv, Nature 297:17-18 (1982) for a description of elements that enhance the activity of eukaryotic promoters. The enhancer can be spliced ​​into the vector at either the 5' or 3' side of the antibody polypeptide coding sequence, so long as enhancement is achieved, but is generally located at a site 5' from the promoter.

[0095] Multispecific multimeric polypeptides

[0117] Certain aspects of the disclosure relate to multimeric polypeptides, e.g., comprising two or more subunits, each comprising two or more polypeptide chains, In some embodiments, one or more of the polypeptide chains, one or more subunits, or one or more multimeric polypeptides of the disclosure are non-native to the host cell.

[0096]

[0118] In some embodiments, the subunits of the present disclosure are monomers of a heterodimer. As used herein, a heterodimer may refer to any polypeptide complex that contains two separate polypeptides or polypeptide complexes that are operably linked. A non-limiting example of a heterodimer is a bispecific or bivalent antibody (i.e., a light chain-heavy chain pair that is operably linked) that is composed of two separate antibody monomers. In this example, folding and assembly of a first heavy chain-light chain pair that recognizes a first antigen produces a first antibody monomer. Folding and assembly of a second heavy chain-light chain pair that recognizes a second antigen produces a second antibody monomer. These monomers can be assembled by any means known in the art (described in more detail below with respect to bispecific antibodies) to form a heterodimer. For further details regarding exemplary heterodimeric antibody formation, see Ridgway JBB et al. 1996 Protein Eng. 9(7):617-621.

[0097]

[0119] In some embodiments, the multimeric polypeptides or subunits of the present disclosure are secreted proteins. As used herein, a secreted protein may refer to any protein that is secreted by a host cell into the host cell periplasm or extracellular environment. A secreted protein may be a protein that is endogenously secreted by a host cell, or a secreted protein may be a protein that is not endogenously secreted by a host cell but is modified in a manner to facilitate its secretion. For example, the presence of a signal sequence, typically found at the N-terminus of a polypeptide, may direct the polypeptide into the secretory pathway for secretion. A number of signal sequences are known in the art and may be useful to facilitate the secretion of secreted proteins or to enable the secretion of proteins that are not naturally secreted by a host cell. See, e.g., Picken et al., Infect. Immun. 42:269-275(1983); Simmons and Yansura, Nature Biotechnology 14:629-634(1996); and Humphreys DP et al. 2000 Protein Expr. Purif. 20(2):252. One non-limiting example of a signal sequence is the heat-stable enterotoxin II (STII) signal sequence.

[0098]

[0120] In some embodiments, the polypeptide chains of the subunits of the present disclosure are linked together by at least one disulfide bond. In some embodiments, the subunits of the multimeric polypeptides of the present disclosure are linked together by at least one disulfide bond. A disulfide bond can refer to any covalent bond linking two thiol groups. Disulfide bonds in polypeptides are typically formed between the thiol groups of cysteine ​​residues. Polypeptide disulfide bonds are known in the art to be important for the folding and assembly of many polypeptides, such as the two-chain proteins of the present disclosure. Polypeptide disulfide bonds can include disulfide bonds between cysteine ​​residues in a single polypeptide chain (i.e., intramolecular or intrachain disulfide bonds). Polypeptide disulfide bonds can also include disulfide bonds between cysteine ​​residues found on separate polypeptide chains (i.e., intermolecular or interchain disulfide bonds).

[0099]

[0121] Disulfide bonds are known in the art to be important for the folding and assembly of antibodies and antibody fragments. Different antibody isotopes, and different subclasses within an isotope, are known to possess different patterns of disulfide bonds. For example, an IgG antibody may contain 12 intrachain disulfide bonds, one interchain disulfide bond between each light chain and its corresponding heavy chain, and 2-11 interchain disulfide bonds between heavy chains, depending on the particular IgG subclass (see Liu H and May K 2012 MAbs. 4(1):17 for a more detailed description). IgM (see, e.g., Wiersma EJ and Shulman MJ 1995 J. Immunol. 154(10):5265), IgE (see, e.g., Helm BA et al. 1991 Eur. J. Immunol. 21(6):1543), IgA (see, e.g., Chintalacharuvu KR et al. 2002 J. Immunol. 169(9):5072), and IgD (see, e.g., Shin SU et al. 1992 Hum. Antibodies Hybridomas 3(2):65) are also known to form disulfide bonds during folding and assembly.

[0100]

[0122] In some embodiments, the multimeric polypeptides of the present disclosure include multispecific antigen-binding proteins. In some embodiments, multispecific antigen-binding proteins bind to two or more epitopes of one, two, or more polypeptides or other antigens. Multispecific antibodies have binding specificities for at least two different epitopes, which usually originate from different antigens. While such molecules usually only bind to two different epitopes (i.e., bispecific antibodies, BsAbs), antibodies with additional specificities, such as trispecific antibodies, are encompassed by this term as used herein. Bispecific antibodies can be prepared as full-length antibodies or antibody fragments (e.g., F(ab')2 bispecific antibodies).

[0101]

[0123] In some embodiments, the multimeric polypeptide of the present disclosure is an antibody. In some embodiments, the subunit of the present disclosure is a half antibody. In some embodiments, the antibody provided herein is a chimeric, human, or humanized antibody. Antibodies or antibody fragments isolated from a human antibody library are considered herein as human antibodies or human antibody fragments. As described below, antibodies are prepared using techniques available in the art for producing antibodies, exemplary methods of which are described in more detail in the following sections. Those skilled in the art will recognize that many of the methods described below can be applied to multimeric polypeptides other than antibodies.

[0102]

[0124] In some embodiments, the subunit of the present disclosure is a monovalent antibody, where the first and second chains correspond to an immunoglobulin heavy chain and an immunoglobulin light chain. As used herein, a monovalent antibody may refer to any polypeptide complex made up of an antibody heavy chain and an antibody light chain that are operatively linked together to form a heavy-light chain pair, where the heavy-light chain pair is not operatively linked to a second heavy-light chain pair. The term "half antibody (hAb)" may be used interchangeably herein.

[0103]

[0125] In some embodiments, the multimeric polypeptide specifically binds to one or more target antigens. In some embodiments, the multimeric polypeptide or subunits of the present disclosure can specifically bind to an antigen. As used herein, the terms "bind", "specifically bind" or "specific to" refer to a measurable and reproducible interaction, such as binding, between a target (i.e., an antibody that determines the presence of a target in the presence of a heterogeneous population of molecules, including biomolecules). For example, an antibody that binds to or specifically binds to a target (which may be an epitope) is an antibody that binds to this target with higher affinity, avidity, more readily, and / or for longer than it binds to other targets. In one embodiment, the extent to which an antibody binds to an unrelated target is less than about 10% of the binding of the antibody to the target, as measured, for example, by radioimmunoassay (RIA). In certain embodiments, an antibody that specifically binds to a target has a dissociation constant (Kd) of 1 μM or less, 100 nM or less, 10 nM or less, 1 nM or less, or 0.1 nM or less. In certain embodiments, the antibody specifically binds to an epitope on a protein that is conserved among proteins from different species. In other embodiments, specific binding can include, but does not require, exclusive binding. In certain embodiments, the antibodies provided herein have an affinity for at least 1 μM, at most 150 nM, at most 100 nM, at most 50 nM, at most 10 nM, at most 1 nM, at most 0.1 nM, at most 0.01 nM, or at most 0.001 nM (e.g., 10 -8 M or less, e.g., 10 -8 M to 10 -13 M, for example 10 -9 M to 10 -13 It has a dissociation constant (Kd) of 1 M.

[0104]

[0126] In one embodiment, Kd is measured by a radiolabeled antigen binding assay (RIA) performed with a Fab version of the antibody of interest and its antigen, as illustrated by the following assay: The solution binding affinity of the Fab for the antigen is determined by the minimum concentration of ( 125I) Fab is equilibrated with labeled antigen, and then bound antigen is measured by capturing it on a plate coated with an anti-Fab antibody (see, e.g., Chen et al., J. Mol. Biol. 293:865-881 (1999)). To establish assay conditions, MICROTITER® multiwell plates (Thermo Scientific) are coated overnight with 5 μg / mL of capturing anti-Fab antibody (Cappel Labs) in 50 mM sodium carbonate (pH 9.6), then blocked with 2% (w / v) bovine serum albumin in PBS for 2-5 hours at room temperature (approximately 23° C.). In a non-adsorbent plate (Nunc #269620), 100 pM or 26 pM [ 125 I] Antigen is mixed with serial dilutions of the Fab of interest. The Fab of interest is then incubated overnight, although incubation may be continued for a longer period (e.g., about 65 hours) to ensure equilibrium is reached. The mixture is then transferred to a capture plate for incubation at room temperature (e.g., 1 hour). The solution is then removed and the plate is washed 8 times with 0.1% polysorbate 20 (TWEEN-20®) in PBS. Once the plate is dry, 150 μL / well of scintillant (MICROSCINT-20®) is added. TM , Packard) and the plate is TOPCOUNT TM Count for 10 minutes on a gamma counter (Packard). Concentrations of each Fab that give 20% or less of maximal binding are selected for use in competitive binding assays.

[0105]

[0127] According to another embodiment, Kd is measured using a surface plasmon resonance assay using a BIACORE®-2000 or BIACORE®-3000 (BIAcore, Inc., Piscataway, NJ) at 25° C. with an immobilized antigen CM5 chip at about 10 response units (RU). Briefly, a carboxymethylated dextran biosensor chip (CM5, BIACORE, Inc.) is activated with N-ethyl-N'-(3-dimethylaminopropyl)-carbodiimide hydrochloride (EDC) and N-hydroxysuccinimide (NHS) according to the supplier's instructions. The antigen is diluted to 5 μg / ml (about 0.2 μM) with 10 mM sodium acetate, pH 4.8, and then injected at a flow rate of 5 μl / min to achieve approximately 10 response units (RU) of coupled protein. After antigen injection, 1 M ethanolamine is injected to block unreacted groups. For kinetic measurements, two-fold serial dilutions of Fab (0.78 nM to 500 nM) were incubated at 25 °C with 0.05% polysorbate 20 (TWEEN-20) at a flow rate of approximately 25 μL / min. TM ) in PBS with detergent (PBST). Association rates (kon) and dissociation rates (koff) are calculated by simultaneously fitting the association and dissociation sensorgrams using a simple 1:1 Langmuir binding model (BIACORE® Evaluation Software version 3.2). The equilibrium dissociation constant (Kd) is calculated as the ratio koff / kon. See, e.g., Chen et al., J. Mol. Biol. 293:865-881 (1999). If the on-rate exceeds 106 M-1 s-1 by the surface plasmon resonance assay described above, the on-rate is measured using a spectrophotometer equipped with stop-flow (Aviv Instruments) or an 8000 series SLM-AMINCO with a stirred cuvette. TMIt can be determined by using a fluorescence quenching technique that measures the increase or decrease in fluorescence emission intensity (excitation = 295 nm, emission = 340 nm, 16 nm band pass) of 20 nM anti-antigen antibody (Fab form) in PBS (pH 7.2) at 25°C in the presence of increasing concentrations of antigen as measured with a spectrometer such as a spectrophotometer (ThermoSpectronic).

[0106]

[0128] In some embodiments, the multimeric polypeptide of the present disclosure comprises a bispecific antibody. In some embodiments, the first and third polypeptide chains are antibody heavy chains and the second and fourth polypeptide chains are antibody light chains. In some embodiments, the first subunit is a first half antibody that binds to a first antigen and the second subunit is a second half antibody that binds to a second antigen. In some embodiments, the first and second antigens are different. In some embodiments, the first and second antigens correspond to different epitopes of the same target. Other antibodies, such as trispecific or tetravalent antibodies, are also contemplated.

[0107]

[0129] Methods for producing bispecific antibodies are known in the art. Traditional production of full-length bispecific antibodies is based on the co-expression of two immunoglobulin heavy-light chain pairs, where these two chains have different specificities (Millstein et al., Nature, 305:537-539 (1983)). Due to the random assortment of immunoglobulin heavy and light chains, these hybridomas (quadromas) may produce a mixture with 10 different antibody molecules, only one of which has the correct bispecific structure. Purification of this correct molecule, usually performed by an affinity chromatography step, is somewhat cumbersome and the product yield is low. Similar procedures are described in WO 93 / 08829 and Traunecker et al., EMBO J., 10:3655-3659 (1991).

[0108]

[0130] In certain embodiments, one or more amino acid modifications can be introduced into the Fc region of an antibody provided herein, thereby generating an Fc region variant. The Fc region variant can comprise a human Fc region sequence (e.g., a human IgG1, IgG2, IgG3, or IgG4 Fc region) that contains an amino acid modification (e.g., a substitution) at one or more amino acid positions.

[0109]

[0131] In certain embodiments, the present disclosure contemplates antibody variants that possess some, but not all, effector functions, making them desirable candidates for applications where antibody half-life in vivo is important, but where certain effector functions (such as complement and ADCC) are unnecessary or deleterious. In vitro and / or in vivo cytotoxicity assays can be performed to confirm reduced / depleted CDC and / or ADCC activity. For example, Fc receptor (FcR) binding assays can be performed to ensure that the antibody lacks FcγR binding (and thus likely lacks ADCC activity) but retains FcRn binding ability. NK cells, the primary cells for mediating ADCC, express only Fc(RIII, whereas monocytes express Fc(RI, Fc(RII, and Fc(RIII. FcR expression on hematopoietic cells is summarized in Table 3, p. 464 of Ravetch and Kinet, Annu. Rev. Immunol., 9:457-492 (1991). Non-limiting examples of in vitro assays to assess ADCC activity of a molecule of interest include those described in U.S. Pat. No. 5,500,362 (see, e.g., Hellstrom, I. et al., Proc. Nat'l Acad. Sci. USA 83:7059-7063 (1986)), and Hellstrom, I. et al., Proc. Nat'l Acad. Sci. USA 83:7063-7064 (1986)). Acad. Sci. USA 82:1499-1502 (1985); 5,821,337 (see Bruggemann, M. et al., J. Exp. Med. 166:1351-1361 (1987)). Alternatively, non-radioactive assay methods may be used, such as ACTI for flow cytometry (Cell Technology, Inc. Mountain View, Calif.). TMNon-radioactive cytotoxicity assays, and CytoTox 96® non-radioactive cytotoxicity assays (Promega, Madison, WI). Useful effector cells for such assays include peripheral blood mononuclear cells (PBMCs) and natural killer (NK) cells. Alternatively, or in addition, the ADCC activity of interest can be assessed in vivo, for example in an animal model such as that disclosed in Clynes et al., Proc. Nat'l Acad. Sci. USA 95:652-656 (1998). A C1q binding assay may be performed to confirm that the antibody lacks CDC activity due to its inability to bind C1q. See, for example, C1q and C3c binding ELISAs in WO 2006 / 029879 and WO 2005 / 100402. To assess complement activation, a CDC assay can be performed (see, e.g., Gazzano-Santoro et al., J. Immunol. Methods 202:163 (1996); Cragg, MS et al., Blood 101:1045-1052 (2003); and Cragg, MS and MJ Glennie, Blood 103:2738-2743 (2004)); determination of FcRn binding and in vivo clearance / half-life can also be performed using methods known in the art (see, e.g., Petkova, SB et al., Int'l. Immunol. 18(12):1759-1769 (2006)).

[0110]

[0132] Antibodies with reduced effector function include those with one or more substitutions at residues 238, 265, 269, 270, 297, 327 and 329 in the Fc region (U.S. Patent No. 6,737,056). Such Fc variants include Fc variants with substitutions at two or more of amino acid positions 265, 269, 270, 297 and 327, including the so-called "DANA" Fc variant in which residues 265 and 297 are substituted with alanine (U.S. Patent No. 7,332,581).

[0111] Certain antibody variants have been described with improved or diminished binding to FcRs (see, e.g., U.S. Pat. No. 6,737,056; WO 2004 / 056312; and Shields et al., J. Biol. Chem. 9(2):6591-6604 (2001)).

[0112]

[0134] In certain embodiments, the antibody variant comprises an Fc region with one or more amino acid substitutions that improve ADCC, e.g., substitutions at Fc region positions 298, 333, and / or 334 (EU numbering of residues). In one exemplary embodiment, the antibody comprises the following amino acid substitutions in its Fc region: S298A, E333A, and K334A.

[0113]

[0135] In some embodiments, modifications are made in the Fc region that result in altered (i.e., either improved or decreased) C1q binding and / or complement dependent cytotoxicity (CDC), e.g., as described in U.S. Pat. No. 6,194,551, WO 99 / 51642, and Idusogie et al. J. Immunol. 164:4178-4184 (2000).

[0114]

[0136] Antibodies with increased half-lives and improved binding to the neonatal Fc receptor (FcRn), which is involved in the transfer of maternal IgG to the fetus (Guyer et al., J. Immunol. 117:587 (1976) and Kim et al., J. Immunol. 24:249 (1994)), are described in U.S. Patent Publication No. 2005 / 0014934 (Hinton et al.). These antibodies comprise an Fc region with one or more substitutions which improve binding of the Fc region to FcRn. Such Fc variants include variants having substitutions at one or more of the following Fc region residues: 238, 256, 265, 272, 286, 303, 305, 307, 311, 312, 317, 340, 356, 360, 362, 376, 378, 380, 382, ​​413, 424, or 434, such as a substitution at Fc region residue 434 (U.S. Pat. No. 7,371,826). See also Duncan & Winter, Nature 322:738-40 (1988), U.S. Pat. No. 5,648,260, U.S. Pat. No. 5,624,821, and WO 94 / 29351 for other examples of Fc region variants.

[0115]

[0137] The antibodies of the present disclosure may be further modified to contain additional nonproteinaceous moieties that are known in the art and readily available, hi certain embodiments, the moieties suitable for derivatization of the antibodies are water soluble polymers.

[0116] Knob-into-hole approach

[0138] One approach known in the art for making bispecific antibodies is the "knob-into-hole" or "protuberance-into-cavity" approach (see, e.g., U.S. Pat. No. 5,731,168). In this approach, two immunoglobulin polypeptides (e.g., heavy chain polypeptides) each contain an interface. An interface of one immunoglobulin polypeptide interacts with a corresponding interface of the other immunoglobulin polypeptide, thereby allowing the association of the two immunoglobulin polypeptides. These interfaces can be engineered such that a "knob" or "protuberance" (these terms may be used interchangeably herein) located at the interface of one immunoglobulin polypeptide corresponds to a "hole" or "cavity" (these terms may be used interchangeably herein) located at the interface of the other immunoglobulin polypeptide. In some embodiments, the hole is of the same or similar size as the knob and is suitably positioned such that when the two interfaces interact, a knob of one interface is positionable within a corresponding hole of the other interface. Without wishing to be bound by theory, this is believed to stabilize the heteromultimer and favor the formation of heteromultimers over other species, e.g., homomultimers. In some embodiments, this approach can be used to promote heteromultimerization of two different immunoglobulin polypeptides to generate bispecific antibodies comprising two immunoglobulin polypeptides with binding specificities for different epitopes.

[0117]

[0139] In some embodiments, knobs can be constructed by replacing small amino acid side chains with larger side chains. In some embodiments, holes can be constructed by replacing large amino acid side chains with smaller side chains. The knobs or holes can be present at the original interface or can be synthetically introduced. For example, knobs or holes can be recombinantly introduced by modifying a nucleic acid sequence encoding the interface and replacing at least one "original" amino acid residue with at least one "import" amino acid residue. Methods for modifying nucleic acid sequences can include standard molecular biology techniques known in the art. The side chain volumes of various amino acid residues are shown in the table below. In some embodiments, the original residues have small side chain volumes (e.g., alanine, asparagine, aspartic acid, glycine, serine, threonine, or valine) and the imported residues to form the knobs are naturally occurring amino acids and can include arginine, phenylalanine, tyrosine, and tryptophan. In some embodiments, the original residue has a large side chain volume (e.g., arginine, phenylalanine, tyrosine, and tryptophan) and the imported residue to form the hole is a naturally occurring amino acid, which may include alanine, serine, threonine, and valine.

[0118]

[0140] In some embodiments, the original residues for forming the knob or hole are identified based on the three-dimensional structure of the heteromultimer. Techniques for obtaining the three-dimensional structure known in the art may include X-ray crystallography and NMR. In some embodiments, the interface is the CH3 domain of an immunoglobulin constant domain. In these embodiments, the CH3 / CH3 interface of IgG1 comprises 16 residues on each domain located on four antiparallel β-strands. Without wishing to be bound by theory, the mutated residues are preferably located on these two central antiparallel β-strands to minimize the risk that the knob may be accommodated by the surrounding solvent rather than a compensatory hole in the partner CH3 domain. In some embodiments, the mutations that form the corresponding knob and hole in the two immunoglobulin polypeptides correspond to one or more pairs provided in the table below.

[0119]

[0141] In certain embodiments, the CH3 and / or CH2 domains of the antibodies of the present disclosure are derived from the IgG4 subtype. In some embodiments, the IgG4 CH3 and / or CH2 domains of the antibodies of the present disclosure may comprise one or more additional mutations, including but not limited to, the S228P mutation (EU numbering).

[0120] TIFF0007658905000003.tif122170 a Molecular weights of amino acids are minus the molecular weight of water. Values ​​from Handbook of Chemistry and Physics, 43rd ed., Cleveland, Chemical Rubber Publishing Co., 1961. b Values ​​from AA Zamyatnin, Prog. Biophys. Mol. Biol. 24:107-123, 1972. c Values ​​from C. Chothia, J. Mol. Biol. 105:1-14, 1975. Accessible surface area is defined in Figures 6-20 of this reference.

[0121]

[0142] In some embodiments, the polypeptide chain of one subunit comprises at least one hole mutation and the polypeptide chain of the other subunit comprises at least one knob mutation, e.g., a first subunit comprises an antibody Fc region comprising at least one hole mutation and a second subunit comprises an antibody Fc region comprising at least one knob mutation, or a first subunit comprises an antibody Fc region comprising at least one knob mutation and a second subunit comprises an antibody Fc region comprising at least one hole mutation.

[0122]

[0143] In some embodiments, the CH3 domain of an antibody of the disclosure is derived from an IgG (e.g., an IgG1 subtype, an IgG2 subtype, an IgG2A subtype, an IgG2B subtype, an IgG3 subtype, or an IgG4 subtype). In some embodiments, the CH3 domain of an antibody of the disclosure may comprise one or more knob or hole mutations (e.g., such as those listed in Table 3 below).

[0123] TIFF0007658905000004.tif63170

[0124]

[0144] In some embodiments, the immunoglobulin polypeptide comprises a CH3 domain comprising one or more amino acid substitutions listed in Table 2 above. In some embodiments, the bispecific antibody comprises a first immunoglobulin polypeptide comprising a CH3 domain comprising one or more amino acid substitutions listed in the left column of Table 3, and a second immunoglobulin polypeptide comprising a CH3 domain comprising one or more corresponding amino acid substitutions listed in the right column of Table 3. In some embodiments, one subunit of a multimeric polypeptide of the disclosure comprises a mutation listed in the left column of a row in Table 3, and another subunit of the multimeric polypeptide comprises a mutation listed in the right column of the same row in Table 3. As a non-limiting example of a knob and hole pair, in some embodiments, the bispecific antibody comprises a first immunoglobulin polypeptide comprising a CH3 domain comprising a T366W mutation, and a second immunoglobulin polypeptide comprising a CH3 domain comprising T366S, L368A, and Y407V mutations. In some embodiments, the at least one knob mutation is selected from the group consisting of T366Y, T366W, T394W, and F405W, numbered according to the EU index for human IgG1. In some embodiments, the at least one hole mutation is selected from the group consisting of F405A, Y407T, Y407A, T366S, L368A, Y407V, and T394S, numbered according to the EU index for human IgG1. In certain embodiments, the knob mutation comprises a T366W substitution and the hole mutation comprises a T366S, L368A, and Y407V substitution (numbered according to the EU index for human IgG1). Further descriptions of knob and hole mutations useful in the methods and cells described herein can be found, for example, in Ridgway JBB et al. 1996 Protein Eng. 9(7):617-621 and at www.imgt.org / IMGTbiotechnology / Knobs-into-holes_IgG.html.

[0125]

[0145] After mutating the DNA as described above, polynucleotides encoding modified immunoglobulin polypeptides having one or more corresponding knob or hole mutations can be expressed and purified using standard recombinant techniques and cell systems known in the art. For example, all four polypeptide chains of a bispecific antibody can be produced and assembled in a single host cell, for example, as described and illustrated in the Examples below.

[0126]

[0146] Bispecific antibodies have been produced using leucine zippers. Kostelny et al., J. Immunol., 148(5):1547-1553 (1992). Leucine zipper peptides from the Fos and Jun proteins were attached to the Fab' portions of two different antibodies by gene fusion. The antibody homodimers were reduced at the hinge region to form monomers and then reoxidized to form the antibody heterodimers. This method can also be used to produce antibody homodimers. The "diabody" technology described by Hollinger et al., Proc. Natl. Acad. Sci. USA 90:6444-6448 (1993) provides an alternative mechanism for making bispecific antibody fragments. These fragments consist of a light chain variable domain (V) separated by a linker that is too short to allow pairing between the two domains on the same chain. L ) connected to a heavy chain variable domain (V H ) in a given fragment. H Domain and V L Domain complementary to another fragment V L Domain and V H The domains are paired together to form two antigen-binding sites. Another method for making bispecific antibody fragments by using single-chain Fv (sFv) dimers has also been reported. See Gruber et al., J. Immunol, 152:5368 (1994).

[0127]

[0147] Another technique for making bispecific antibody fragments is the "bispecific T cell engager" or BiTE® approach (see, for example, WO 2004 / 106381, WO 2005 / 061547, WO 2007 / 042261, and WO 2008 / 119567). This approach utilizes two antibody variable domains arranged on a single polypeptide. For example, the single polypeptide chain may comprise a variable heavy chain (V H ) and variable light chain (V L The single polypeptide comprises two single chain Fv (scFv) fragments each having a CD3 domain, which are separated by a polypeptide linker of sufficient length to allow intramolecular association between the two domains. The single polypeptide further comprises a polypeptide spacer sequence between the two scFv fragments. Each scFv recognizes a different epitope, and these epitopes may be specific for different cell types, such that cells of two different cell types are brought into close proximity or tethered when each scFv engages its cognate epitope. One particular embodiment of this approach comprises an scFv that recognizes a cell surface antigen expressed by an immune cell, e.g., a CD3 polypeptide on a T cell, which is joined to another scFv that recognizes a cell surface antigen expressed by a target cell, such as a malignant or tumor cell.

[0128]

[0148] Antibodies with more than two valencies are contemplated. For example, trispecific antibodies can be prepared. Tuft et al., J. Immunol. 147:60 (1991).

[0129]

[0149] In some embodiments, the antibody of the present disclosure is a single domain antibody. A single domain antibody is a single polypeptide chain that includes all or a portion of the heavy chain variable domain or all or a portion of the light chain variable domain of an antibody. In certain embodiments, a single domain antibody is a human single domain antibody (Domantis, Inc., Waltham, Mass.; see, e.g., U.S. Pat. No. 6,248,516 B1). In one embodiment, a single domain antibody consists of all or a portion of the heavy chain variable domain of an antibody.

[0130] Mutations that promote selective light-chain-heavy-chain pairing

[0150] In some embodiments, the multimeric polypeptides of the present disclosure comprise one or more mutations that promote selective association between the antibody light chain and the antibody heavy chain. Exemplary amino acid substitutions that promote selective association between the antibody heavy chain and the antibody light chain useful in the methods and cells described herein can be found in WO 2016 / 172485. In some embodiments, the antibody heavy chain of the subunit comprises one or more mutations that promote selective association, and the antibody light chain of the subunit comprises one or more mutations that promote selective association in combination with the heavy chain. In some embodiments, both half antibodies of the bispecific antibody comprise one or more mutations that promote selective association between the antibody light chain and the antibody heavy chain (e.g., one heavy chain has a positively charged mutation and its corresponding light chain has a negatively charged mutation, while the other heavy chain has a negatively charged mutation and its corresponding light chain has a positively charged mutation, or vice versa). In some embodiments, only one of the half antibodies of the bispecific antibody contains one or more mutations that promote selective association of an antibody light chain with an antibody heavy chain (e.g., one heavy chain contains a positively charged mutation and its corresponding light chain contains a negatively charged mutation, or vice versa).

[0131]

[0151] In some embodiments, the antibody heavy chain (e.g., CH1 domain) comprises an amino acid substitution at S183 and the antibody light chain (e.g., CL domain) comprises an amino acid substitution at V133 (numbered according to the EU index). In some embodiments, the S183 substitution is selected from the group consisting of S183A, S183T, S183V, S183Y, S183F, S183H, S183N, S183D, S183E, S183R, and S183K. In some embodiments, the V133 substitution is selected from the group consisting of V133E, V133S, V133L, V133W, V133K, V133R, and V133D. In some embodiments, the amino acid substitution at S183 results in a positively charged residue and the amino acid substitution at V133 results in a negatively charged residue. In some embodiments, the amino acid substitution at S183 results in a negatively charged residue and the amino acid substitution at V133 results in a positively charged residue.

[0132]

[0152] Mutations that promote selective association of heavy / light chain pairs can be combined with mutations that promote selective association of heavy chains (e.g., knob and hole mutations). In some embodiments, a multimeric polypeptide (e.g., a bispecific antibody) comprises a set of knob and hole mutations and one or more mutations that promote selective association of an antibody light chain with an antibody heavy chain. Advantageously, this promotes proper heavy / light chain association in one or both half antibodies and proper assembly of the bispecific antibody (e.g., as opposed to forming dimers of each half antibody), resulting in fewer mispaired by-products. In some embodiments, a polypeptide chain of the present disclosure (e.g., an antibody heavy chain of a bispecific antibody or half antibody) comprises one or more knob or hole mutations and one or more mutations that promote selective association of an antibody light chain with an antibody heavy chain. For example, in some embodiments, a first antibody heavy chain of the bispecific antibody comprises a mutation that confers a negative charge (e.g., an S183E mutation) and a hole mutation, a first antibody light chain comprises a mutation that confers a positive charge (e.g., a V133K mutation), a second antibody heavy chain of the bispecific antibody comprises a mutation that confers a positive charge (e.g., an S183K mutation) and a knob mutation, and a second antibody light chain comprises a mutation that confers a negative charge (e.g., a V133E mutation).

[0133] Multimeric Polypeptides and Compositions of Multimeric Polypeptides

[0153] Further provided are multimeric polypeptides and compositions of multimeric polypeptides or multimeric polypeptides produced by the methods described herein. Advantageously, the present disclosure demonstrates that the methods described herein allow for improved production of multimeric polypeptides with higher yields of desired multimeric polypeptides and / or fewer impurities such as mismatched by-products. Thus, in some embodiments, the multimeric polypeptides and multimeric polypeptide compositions produced by the methods of the present disclosure contain fewer mismatched by-products compared to the multimeric polypeptides and multimeric polypeptide compositions produced by existing techniques (e.g., compared to expression in which one or more, two or more, three or more, or four of the polynucleotides encoding the polypeptide chains of the multimeric polypeptides contain an open reading frame operably linked to a natural or unmodified translation initiation sequence, or expression in which each of the polynucleotides encoding the polypeptide chains of the multimeric polypeptides contains the same translation initiation sequence). In some embodiments, the multimeric polypeptides and multimeric polypeptide compositions produced by the methods of the present disclosure contain a higher ratio of multimeric polypeptides to mispaired by-products compared to the multimeric polypeptides and multimeric polypeptide compositions produced by existing techniques (e.g., compared to expression in which one or more, two or more, three or more, or four of the polynucleotides encoding the polypeptide chains of the multimeric polypeptide contain an open reading frame operably linked to a native or unmodified translation initiation sequence, or expression in which each of the polynucleotides encoding the polypeptide chains of the multimeric polypeptide contains the same translation initiation sequence). In some embodiments, all of the polypeptide chains of the multimeric polypeptide are produced from a single eukaryotic (e.g., mammalian) cell. In some embodiments, the multimeric polypeptide is a bispecific antibody.

[0134] Production of multimeric proteins

[0154] Host cells (e.g., those described in Section III below) are transformed with one or more polynucleotides or vectors (e.g., expression vectors) and cultured in conventional nutrient media modified as appropriate for inducing promoters, selecting transformants, or amplifying genes encoding the desired sequences.

[0135]

[0155] The host cells used to produce the desired multimeric polypeptides of the present disclosure or subunits thereof may be cultured in a variety of media. Commercially available media such as Ham's F10 (Sigma), Minimum Essential Medium ((MEM), (Sigma), RPMI-1640 (Sigma), and Dulbecco's Modified Eagle's Medium ((DMEM), Sigma) are suitable for culturing the host cells. In addition, see Ham et al., Meth. Enz. 58:44(1979); Barnes et al., Anal. Any of the media described in Biochem. 102:255 (1980), U.S. Patent Nos. 4,767,704; 4,657,866; 4,927,762; 4,560,655; or 5,122,469; WO 90 / 03430; WO 87 / 00195; or U.S. Patent Rep. 30,985 may be used as a culture medium for the host cells. Any of these media may contain, as necessary, hormones and / or other growth factors (such as insulin, transferrin, or epidermal growth factor), salts (such as sodium chloride, calcium, magnesium, and phosphates), buffers (such as HEPES), nucleotides (such as adenosine and thymidine), antibiotics (such as GENTAMYCIN), ... insulin, transferrin, or epidermal growth factor), salts (such as insulin, transferrin, or epidermal growth factor), salts (such as insulin, transferrin, or epidermal growth factor), salts (such as insulin, transferrin, or epidermal growth factor), salts (such as insulin, transferrin, or epidermal growth factor), salts (such as insulin, transferrin, or epidermal growth factor), salts (such as insulin, transferrin, or epi TM The culture conditions may include, for example, nutrients (such as drugs), trace elements (defined as inorganic compounds usually present at final densities in the micromolar range), and glucose or an equivalent energy source. Any other necessary supplements may also be included at appropriate concentrations known to those of skill in the art. Culture conditions, such as temperature, pH, etc., will be those already used with the host cell selected for expression and will be apparent to those of skill in the art.

[0136]

[0156] Methods for recovering the multimeric polypeptides of the present disclosure are described herein. When using recombinant techniques, the multimeric polypeptides of the present disclosure or subunits thereof are produced intracellularly or secreted directly into the medium. If the polypeptides are produced intracellularly, as a first step, particulate debris, either host cells or lysed fragments, are removed, for example, by centrifugation or ultrafiltration. If the polypeptides are secreted into the medium, the supernatant from such expression systems is generally first concentrated using a commercially available protein concentration filter, for example, an Amicon or Millipore Pellicon ultrafiltration device. Protease inhibitors such as PMSF may be included in any of the above steps to inhibit proteolysis, and antibiotics may be included to prevent the growth of adventitious contaminants.

[0137]

[0157] Polypeptide compositions prepared from cells can be purified using, for example, hydroxylapatite chromatography, gel electrophoresis, dialysis, and affinity chromatography, with affinity chromatography being the preferred purification technique. The suitability of protein A as an affinity ligand depends on the species and isotype of any immunoglobulin Fc domain present in the antibody. Protein A can be used to purify antibodies based on human gamma 1, gamma 2, or gamma 4 heavy chains (Lindmark et al., J. Immunol. Meth. 62:1-13 (1983)). Protein G is recommended for all mouse isotypes and human gamma 3 (Guss et al., EMBO J. 5:15671575 (1986)). The matrix to which the affinity ligand is attached is most often agarose, although other matrices are available. Mechanically stable matrices such as controlled pore glass or poly(styrenedivinyl)benzene allow for faster flow rates and shorter processing times than can be achieved with agarose. If the antibody contains a CH3 domain, Bakerbond ABX TMResins (JT Baker, Phillipsburg, NJ) are useful for purification. Other techniques for protein purification include fractionation on ion exchange columns, ethanol precipitation, reverse phase HPLC, chromatography on silica, chromatography on heparin, SEPHAROSE on anion or cation exchange resins (such as polyaspartic acid columns). TM Chromatography, chromatofocusing, SDS-PAGE, and ammonium sulfate precipitation are also available depending on the antibody recovered.

[0138]

[0158] After any preliminary purification step(s), the mixture containing the polypeptide of interest and contaminants may be subjected to low pH hydrophobic interaction chromatography, preferably performed at low salt concentration (e.g., about 0-0.25 M salt), using an elution buffer at a pH of about 2.5-4.5. Production of antigen-binding polypeptides may alternatively or additionally (to any of the specific methods described above) include dialyzing a solution containing the mixture of polypeptides.

[0139]

[0159] In one embodiment, the multimeric polypeptides of the present disclosure or subunits thereof produced herein are further purified to obtain substantially homogenous preparations for further assays and uses. Standard protein purification methods known in the art can be used. The following procedures are examples of suitable purification procedures: fractionation on immunoaffinity or ion exchange columns, ethanol precipitation, reverse phase HPLC, chromatography on silica or cation exchange resins (such as DEAE), chromatofocusing, SDS-PAGE, ammonium sulfate precipitation, and gel filtration, for example using Sephadex G-75.

[0140]

[0160] The purification of the multimeric polypeptide can be carried out using known chromatographic techniques, including, for example, protein A or protein G column chromatography. In one embodiment, the multimeric polypeptide of interest can be recovered from the solid phase of the column by elution in a solution containing a chaotropic agent or a mild detergent. Exemplary chaotropic agents and mild detergents include, but are not limited to, guanidine-HCI, urea, lithium perclorate, arginine, histidine, SDS (sodium dodecyl sulfate), Tween, Triton, and NP-40, all of which are commercially available.

[0141]

[0161] In one embodiment, Protein A immobilized on a solid phase is used, for example, for immunoaffinity purification of antigen-binding polypeptides of the invention. Protein A is a 41 kD cell wall protein from Staphylococcus aureus that binds with high affinity to the Fc region of antigen-binding polypeptides. Lindmark et al. (1983) J. Immunol. Meth. 62:1-13. The solid phase on which Protein A is immobilized is preferably a column comprising a glass or silica surface, more preferably a controlled pore glass column or a silicic acid column. In some applications, the column is coated with a reagent such as glycerol to prevent non-specific adhesion of contaminants.

[0142]

[0162] As a first step of purification, the preparation obtained from the cell culture as described above is applied to a Protein A-immobilized solid phase to allow specific binding of the antigen-binding polypeptide of interest to Protein A. The solid phase is then washed to remove contaminants non-specifically bound to the solid phase. The antigen-binding polypeptide (e.g., an antibody, etc.) is recovered from the solid phase by elution.

[0143] Screening Methods

[0163] Also provided is a method for identifying a combination of translation initiation sequences for expressing a multimeric polypeptide in a eukaryotic host cell of the present disclosure. In some embodiments, the method includes providing a library comprising a plurality of eukaryotic host cells, culturing the library of eukaryotic host cells under conditions suitable for expression of the multimeric polypeptide by the plurality of eukaryotic host cells, measuring the amount of multimeric polypeptide produced by a single eukaryotic host cell of the plurality of eukaryotic host cells or a clone of a single eukaryotic host cell of the plurality of eukaryotic host cells, and identifying the first, second, third, and fourth translation initiation sequences of one or more of the single eukaryotic host cells of the plurality of eukaryotic host cells or the clone of a single eukaryotic host cell of the plurality of eukaryotic host cells that produce the multimeric polypeptide. Examples of such methods are described and illustrated below.

[0144]

[0164] In some embodiments, the library of the present disclosure comprises a plurality of host cells of the present disclosure. In some embodiments, two or more, or all, of the plurality of host cells comprise a set of polynucleotides necessary to encode each polypeptide chain of a subunit or multimeric polypeptide described herein. For example, in some embodiments, a plurality or all of the host cells in the library comprise a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding a first polypeptide chain of a multimeric polypeptide of the present disclosure, a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding a second polypeptide chain of a multimeric polypeptide of the present disclosure, a third polynucleotide comprising a third translation initiation sequence operably linked to a third open reading frame encoding a third polypeptide chain of a multimeric polypeptide of the present disclosure, and a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to a fourth open reading frame encoding a fourth polypeptide chain of a multimeric polypeptide of the present disclosure. In some embodiments, the plurality of host cells or library comprises a plurality of combinations of the first, second, third, and fourth translation initiation sequences. Thus, screening of the library can be used to identify a combination of translation initiation sequences that results in a desired property (e.g., high levels of expression or low levels of expression of a multimeric polypeptide, such as a mismatched by-product). The present disclosure demonstrates that expression of a multimeric polypeptide can be improved by adjusting the strength of the translation initiation sequences operably linked to the polynucleotides encoding one or more of the constituent polypeptide chains.

[0145]

[0165] In some embodiments, one or more of the translation initiation sequences operably linked to the open reading frame in each host cell of the plurality of host cells comprises the sequence (5' to 3') NNNNNATGNGA, where N is C, G, A, or T / U (SEQ ID NO:1).

[0146]

[0166] Various methods described herein can be used to measure the amount of multimeric polypeptides produced by one or more host cells or clonal populations of host cells of the present disclosure. For example, transient or stable transfectants can be cultured and the production of multimeric polypeptides can be measured. Fed-batch or perfusion cultures can be performed and the titer of the product in the cell medium or on the cell surface can be measured. For surface expression, cells can be stained with antibodies to detect the product and analyzed, for example, by flow cytometry. The quality and / or purity of the product can be assessed, for example, using electrophoresis and / or mass spectrometry.

[0147]

[0167] Antibodies of the present disclosure can be isolated by screening combinatorial libraries for antibodies with the desired activity(ies). A variety of methods are known in the art for generating phage display libraries and screening such libraries for antibodies with the desired binding characteristics, such as, for example, the method described in Example 3. Additional methods are reviewed, for example, in Hoogenboom et al., Methods in Molecular Biology, 178:1-37 (O'Brien et al., eds., Human Press, Totowa, NJ, 2001), and described in, for example, McCafferty et al., Nature, 348:552-554; Clackson et al., Nature, 352:624-628 (1991); Marks et al., J. Mol. Biol., 222:581-597 (1992); Marks and Bradbury, Methods in Molecular Biology, 248:161-175 (Lo, eds., Human Press, Totowa, NJ, 2001). Press, Totowa, NJ, 2003), Sidhu et al., J. Mol. Biol., 338(2):299-310 (2004), Lee et al., J. Mol. Biol., 340(5):1073-1093 (2004), Fellowes, Proc. Natl. Acad. Sci. USA, 101(34):12467-12472 (2004), and Lee et al., J. Immunol. Methods, 284(1-2):119-132 (2004).

[0148]

[0168] In certain phage display methods, repertoires of VH and VL genes can be cloned separately by polymerase chain reaction (PCR), randomly recombined in phage libraries, and then screened for antigen-binding phage, as described by Winter et al., Ann. Rev. Immunol., vol. 12:433-455 (1994). Phages typically display antibody fragments, either as single-chain Fv (scFv) fragments or as Fab fragments. Libraries from immunized sources provide high affinity antibodies to the immunogen without the need to construct hybridomas. Alternatively, naive repertoires can be cloned (e.g., from humans) without immunization to provide a single source of antibodies against a wide range of non-self and even self antigens, as described by Griffiths et al., EMBO J, 12:725-734 (1993). Finally, natural libraries can also be generated synthetically by cloning unrearranged V gene segments from stem cells and using PCR primers containing random sequences to encode the hypervariable CDR3 regions and perform rearrangement in vitro, as described in Hoogenboom and Winter, J. Mol. Biol. 227:381-388 (1992). Patent publications describing human antibody phage libraries include, for example, U.S. Pat. No. 5,750,373, and U.S. Patent Publication Nos. 2005 / 0079574, 2005 / 0119455, 2005 / 0266000, 2007 / 0117126, 2007 / 0160598, 2007 / 0237764, 2007 / 0292936, and 2009 / 0002360.

[0149] III.Cells

[0169] Also provided herein are recombinant eukaryotic host cells useful for producing multimeric polypeptides, for example according to the methods described in Section II. In some embodiments, the cells comprise a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding a first polypeptide chain of a multimeric polypeptide of the disclosure, a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding a second polypeptide chain of a multimeric polypeptide of the disclosure, a third polynucleotide comprising a third translation initiation sequence operably linked to a third open reading frame encoding a third polypeptide chain of a multimeric polypeptide of the disclosure, and a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to a fourth open reading frame encoding a fourth polypeptide chain of a multimeric polypeptide of the disclosure. In some embodiments, the first subunit is expressed at a lower level than the second subunit when each subunit is expressed individually in a recombinant eukaryotic host cell, and one or both of the first translation initiation sequence and the second translation initiation sequence are weaker than one or both of the third translation initiation sequence and the fourth translation initiation sequence. In some embodiments, the multimeric polypeptide is non-native to the recombinant eukaryotic host cell. In some embodiments, the recombinant eukaryotic host cell is an isolated recombinant eukaryotic host cell.

[0150]

[0170] In some embodiments, the host cell of the disclosure is a eukaryotic cell. In some embodiments, the host cell of the disclosure is a mammalian cell.

[0151]

[0171] Suitable host cells include higher eukaryotic cells as described herein, including vertebrate host cells. Propagation of vertebrate cells in culture (tissue culture) has become a routine procedure. Examples of useful mammalian host cell lines are monkey kidney CV1 line transformed by SV40 (COS-7, ATCC CRL 1651); human embryonic kidney line (293 or 293 cells subcloned for growth in suspension culture, Graham et al., J. Gen Virol. 36:59(1977)); baby hamster kidney cells (BHK, ATCC CCL 77:4216(10); Chinese hamster ovary cells / -DHFR (CHO, Urlaub et al., Proc. Natl. Acad. Sci. USA 77:4216(1980)); mouse Sertoli cells (TTM4, Mather, Biol. Reprod. 23:243-251(1980)); monkey kidney cells (CV1 ATCC CCL 70); African green monkey kidney cells (VERO-76, ATCC CRL-1587); human cervical carcinoma cells (HELA, ATCC CCL 2); canine kidney cells (MDCK, ATCC CCL 34); buffalo rat hepatocytes (BRL 3A, ATCC CRL 1442); human lung cells (W138, ATCC CCL 75); human hepatocytes (Hep G2, HB 8065); mouse mammary tumor (MMT 060562, ATCC CCL51); TRI cells (Mather et al., Annals NY Acad. Sci. 383:44-68(1982)); MRC5 cells; FS4 cells; and a human hepatoma line (HepG2).

[0152]

[0172] In some embodiments, the host cell of the disclosure is a Chinese Hamster Ovary (CHO) cell or cell line. In some embodiments, the host cell of the disclosure is a CHO-K1 cell line (e.g., ATCC catalog number CCL-61 TM(See, e.g., Urlaub, G. and Chasin, LA (1980) Proc. Natl. Acad. Sci. 77:4216-4220). In some embodiments, the host cells of the disclosure are CHO cells deficient in dihydrofolate reductase (DHFR) activity. (See, e.g., Urlaub, G. and Chasin, LA (1980) Proc. Natl. Acad. Sci. 77:4216-4220.)

[0153]

[0173] In some embodiments, all of the polypeptide chains of a multimeric polypeptide are produced from a single eukaryotic (e.g., mammalian) cell. Advantageously, this eliminates the need for post-production assembly of the multimeric polypeptide (e.g., a bispecific antibody), as occurs when two subunits (e.g., half antibodies) are produced and then assembled in vitro.

[0154]

[0174] Any of the promoters and / or translation initiation sequences described herein can find use in the host cells of the disclosure.

[0155]

[0175] In some embodiments, one or more polynucleotides encoding the polypeptide chains of the subunits of a multimeric polypeptide of the present disclosure are present on a chromosome of a host cell of the present disclosure (e.g., via integration). In some embodiments, each of the polynucleotides encoding the polypeptide chains of the subunits of a multimeric polypeptide of the present disclosure are present on a chromosome of a host cell of the present disclosure (e.g., via integration). In some embodiments, each of the polynucleotides encoding the polypeptide chains of the subunits of a multimeric polypeptide of the present disclosure are present on the same chromosome or chromosomal locus of a host cell of the present disclosure (e.g., via integration).

[0156]

[0176] Stable integration or stable transfection of the polynucleotides of the present disclosure into the host cell chromosome can provide a stable production cell line for producing the multimeric polypeptides of the present disclosure. Various approaches suitable for integrating polynucleotides into the host cell genome are known in the art, including random integration or site-specific integration (e.g., the "landing pad" approach). See, for example, Zhao, M. et al. (2018) Appl. Microbiol. Biotechnol. 102:6105-6117; Lee, JS et al. (2015) Sci. Rep. 5:8572; and Gaidukov, L. et al. (2018) Nucleic Acids Res. 46:4072-4086.

[0157]

[0177] In some embodiments, one or more polynucleotides encoding the polypeptide chains of the subunits of the multimeric polypeptides of the present disclosure are maintained as extrachromosomal polynucleotides in the host cell. In some embodiments, each of the polynucleotides encoding the polypeptide chains of the subunits of the multimeric polypeptides of the present disclosure are maintained as extrachromosomal polynucleotides in the host cell. In some embodiments, one or more polynucleotides encoding the polypeptide chains of the subunits of the multimeric polypeptides of the present disclosure are present in a vector (e.g., an expression vector). In some embodiments, each of the polynucleotides encoding the polypeptide chains of the subunits of the multimeric polypeptides of the present disclosure are present in one or more vectors (e.g., an expression vector).

[0158]

[0178] Vector components generally include, but are not limited to, one or more of the following: a signal sequence, an origin of replication, one or more marker genes, an enhancer element, a promoter, and a transcription termination sequence.

[0159]

[0179] Vectors for use in eukaryotic host cells may also contain a signal sequence or other polypeptide with a specific cleavage site at the N-terminus of the mature protein or polypeptide of interest. The heterologous signal sequence preferably selected is one that is recognized and processed (i.e., cleaved by a signal peptidase) by the host cell. In mammalian cell expression, mammalian signal sequences and viral secretory leaders, such as the herpes simplex gD signal, are available. The DNA of such a precursor region is ligated in reading frame to DNA encoding the desired antigen-binding polypeptide(s) (e.g., an antibody).

[0160]

[0180] Generally, the origin of replication component is not needed for mammalian expression vectors For example, the SV40 origin may typically be used only because it contains the early promoter.

[0161]

[0181] Any of the promoters and / or translation initiation sequences described herein can find use in the vectors of the disclosure.

[0162]

[0182] Expression and cloning vectors can contain a selection gene, also known as a selectable marker. Typical selection genes encode (a) a protein that confers resistance to antibiotics or other toxins, such as ampicillin, neomycin, methotrexate, or tetracycline, (b) a protein that complements an auxotrophic deficiency, if relevant, or (c) a protein that supplies a vital nutrient not available from complex media.

[0163]

[0183] One example of a selection scheme utilizes a drug to arrest the growth of the host cell. Cells successfully transformed with a heterologous gene produce a protein that confers drug resistance and therefore survive the selection regime. Examples of such dominant selection use the drugs neomycin, mycophenolic acid, and hygromycin.

[0164]

[0184] Other examples of suitable selectable markers for mammalian cells are those that enable the identification of cells competent to take up the antibody nucleic acid, such as DHFR, thymidine kinase, metallothionein-I and -II, preferably a primate metallothionein gene, adenosine deaminase, ornithine decarboxylase, and the like.

[0165]

[0185] For example, cells transformed with the DHFR selection gene are first identified by culturing all of the transformants in a culture medium containing methotrexate (Mtx), a competitive antagonist of DHFR.When wild-type DHFR is used, an appropriate host cell is a Chinese hamster ovary (CHO) cell line deficient in DHFR activity (e.g., ATCC CRL-9096).

[0166]

[0186] Alternatively, host cells transformed or co-transformed with a DNA sequence encoding a polypeptide chain of the disclosure, a wild-type DHFR protein, and another selectable marker, such as an aminoglycoside 3'-phosphotransferase (APH) (specifically a wild-type host containing endogenous DHFR) can be selected by growing the cells in medium containing a selection agent for the selectable marker, such as an aminoglycoside antibiotic, e.g., kanamycin, neomycin, or G418. See, e.g., U.S. Patent No. 4,965,199.

[0167]

[0187] Expression vectors used in eukaryotic host cells typically also contain sequences necessary for transcription termination and stabilization of the mRNA. Such sequences are commonly available from the 5' and, occasionally, 3' untranslated regions of eukaryotic or viral DNA or cDNA. These regions contain nucleotide segments transcribed as polyadenylated fragments in the untranslated portion of the mRNA encoding the antibody. One useful transcription termination component is the bovine growth hormone polyadenylation region. See WO 94 / 11026 and the expression vectors disclosed therein.

[0168]

[0188] Host cells are transformed with the above-described expression or cloning vectors for production of the desired polypeptide(s) (e.g., multimeric polypeptides, etc.) and cultured in conventional nutrient media modified as appropriate for inducing promoters, selecting transformants, or amplifying genes encoding the desired sequences.

[0169] IV. KITS AND ARTICLES

[0189] Further provided herein are kits or articles of manufacture useful, for example, for expressing multimeric polypeptides.

[0170]

[0190] In some embodiments, the kit comprises a set of polynucleotides encoding each of the polypeptide chains and / or subunits of the multimeric polypeptide. For example, in some embodiments, the kit comprises a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding a first polypeptide chain of the multimeric polypeptide of the disclosure, a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding a second polypeptide chain of the multimeric polypeptide of the disclosure, a third polynucleotide comprising a third translation initiation sequence operably linked to a third open reading frame encoding a third polypeptide chain of the multimeric polypeptide of the disclosure, and a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to a fourth open reading frame encoding a fourth polypeptide chain of the multimeric polypeptide of the disclosure. In some embodiments, one or more of the first, second, third, and fourth translation initiation sequences are not operably linked to their respective open reading frames when the respective open reading frames are present in the naturally occurring host cell genome.

[0171]

[0191] In some embodiments, each of the polynucleotides in the kit of the present disclosure is part of one or more expression vectors of the present disclosure.

[0172]

[0192] In some embodiments, each of the polynucleotides in the kit of the present disclosure is operably linked to a promoter of the present disclosure. For example, in some embodiments, each of the polynucleotides is operably linked to a different promoter. In some embodiments, two or more of the polynucleotides are operably linked to the same promoter. For example, when the multimeric polypeptide is a bispecific antibody, the polynucleotides encoding the open reading frames of the heavy and light chains of each antibody can be operably linked to the same promoter. In some embodiments, the "same promoter" refers to the same physical polynucleotide. In some embodiments, the "same promoter" refers to physically different polynucleotides that share the same promoter sequence.

[0173]

[0193] In some embodiments, the kit or article of manufacture further comprises instructions for use of the set of polynucleotides for production of a subunit or multimeric polypeptide of the present disclosure, e.g., according to any of the methods described in Section II above, or using any of the host cells described in Section III above. EXAMPLES

[0174]

[0194] The present disclosure will be more fully understood by referring to the following examples. However, the examples should not be interpreted as limiting the scope of the present disclosure. It is understood that the examples and embodiments described herein are for illustrative purposes only, and various modifications or changes taking into account thereof are suggested to those skilled in the art, and should be included within the spirit and scope of the present specification and the scope of the appended claims.

[0175] Example 1: Modulation of protein production by Kozak sequence variants in transient transfection of mammalian cells

[0195] Currently, there are few systematic approaches to precisely control the translational level of recombinant proteins in mammalian cells. The region where translation begins to synthesize a polypeptide is called the translation initiation site (TIS). A translation initiation site (TIS) consists of an initiation codon and its adjacent bases. In eukaryotes, translation initiation usually follows a scanning mechanism model (Kozak M. Cell. 1978;15(4):1109-23), which assumes a ribosomal preinitiation complex consisting of the small 40S ribosomal subunit, Met-tRNA, eIF2-GTP, eIF1, eIF1A, eIF3, and eIF5, which binds at the 5' end of the mRNA and proceeds linearly in the 3' direction in search of an initiation codon. After positioning the small 40S ribosomal subunit in the start codon, the initiation factors dissociate and the large subunit binds to the small 40S ribosomal subunit to form a ribosomal complex and translation begins (Nanda JS, Saini AK, Munoz AM, Hinnebusch AG, Lorsch JR. J Biol Chem. 2013;288(8):5316-29; Pestova TV, Kolupaeva VG. Genes Dev. 2002;16(22):2906-22). Initiation is not necessarily restricted to the start codon (AUG) closest to the 5' end. If the first AUG codon occurs under optimal conditions, the ribosomal complex will initiate translation, but if the translation initiation site (TIS) around the first AUG triplet is suboptimal, some 40S subunits will bypass the site and initiate further downstream.Thus, cells can control protein translation levels by adjusting the TIS, and thus the sequence around the start codon plays a key role in enhancing initiation and subsequent translation efficiency (Kozak M. J Biol Chem. 1991;266(30):19867-70; Kozak M. J Cell Biol. 1991;115(4):887-903; Sonenberg N, Hinnebusch AG. Cell. 2009;136(4):731-45; Ivanov IP, Loughran G, Sachs MS, Atkins JF. Proc Natl Acad Sci US A. 2010;107(42):18056-60).

[0176]

[0196] Kozak is the CCRCC AUGreported that G (purine, R=A or G; start codon underlined) is a highly efficient mammalian TIS (Kozak M. Nucleic Acids Res. 1981;9(20):5233-52). Within that sequence, the purine at position -3 (three nucleotides upstream of the AUG codon) is the most highly conserved in vertebrate messenger RNAs (Kozak M. Nucleic Acids Res. 1987;15(20):8125-48). Point mutation studies provide evidence for the importance of A or G at position -3, and G at position +4 (immediately following the AUG codon) as important for optimal translation efficiency (Kozak M. EMBO J. 1997;16(9):2482-92). The Kozak consensus sequence varies in length and nucleotide composition between species, but is conserved in most genes within a species. Not surprisingly, point mutations in Kozak sequences affect translation initiation in both higher eukaryotes (Kozak M. Cell. 1986;44(2):283-92) and lower eukaryotes (Dvir S, Velten L, Sharon E, Zeevi D, Carey LB, Weinberger A, et al. Proc Natl Acad Sci US A. 2013;110(30):E2792-801) and have been linked to the development of human diseases such as cancer and metabolic disorders (Sonenberg N, Hinnebusch AG. Cell. 2009;136(4):731-45;Mohan RA, van Engelen K, Stefanovic S, Barnett P, Ilgun A, Baars MJ, et al. Am J Med Genet A. 2014;164A(11):2732-8).

[0177]

[0197] To investigate whether variants of the Kozak sequence could modulate protein production in industrialized producer cell lines, we varied the −3 position upstream of the start codon of the ORF encoding the Fc fusion protein and combinations of the −3, −2, and −1 positions.

[0178] method Cell lines and cell culture

[0198] CHO cell lines were used for transient transfection assays in plates and for constitutive antibody production cell lines via targeted integration. CHO cells were cultured in DMEM / F12-based medium in 125 mL shake flasks at 150 rpm, 37 °C and 5% CO2. 3 × 10 5 Cells were passaged at a seeding density of 100 cells / ml.

[0179] Transient transfection

[0199] CHO cells were transformed with the transient transfection plasmids using Lipofectamine 2000 CD according to the manufacturer's recommendations (Invitrogen, Carlsbad, CA). DNA of high and low expressing antibodies was also carried as an internal control. Briefly, DNA vector (2 μg) in 500 μl medium plus 10 μl Lipofectamine was incubated at room temperature for 30 min to promote complex formation. Transfection complexes were transferred to 96-well plates (Falcon®) containing exponentially growing CHO cells in DMEM / F12-based medium plus 5% DFBS, resulting in a total working volume of approximately 2.5 ml. The transformed cells were cultured at 37°C, 5% CO2, and 80% humidity. 24 hours after transfection, the cell culture medium was replaced with production medium, and then the cells were cultured at 33°C. Each transfection was performed in two biological replicates. For determination of antibody concentrations, supernatant samples were harvested from cultures 48 hours after transfection and assayed in duplicate by HTRF (homogeneous time-resolved FRET) assay.

[0180] Homogeneous time-resolved FRET (HTRF) assay

[0200] For HTRF assays, see Degorce, F., et al. (2009) Curr Chem Genomics 3:22-32.

[0181] Design and construction of Kozak sequence variants

[0201] Based on the nucleotide frequencies around vertebrate translation start sites as determined by Kozak (Kozak M. Nucleic Acids Res. 1987;15(20):8125-48), low and intermediate frequency variants at these positions were designed as shown in FIG. 1. DNA for the Kozak sequence variants of Fc fusion protein A or B was synthesized by Genewiz, Inc. The variants were cloned in a transient transfection vector characteristic of Genentech using an ampicillin resistance marker under the transcriptional control of the CMV promoter. Sequence verification was performed using universal forward and universal reverse primers.

[0182] DNA library design

[0202] Fc fusion protein B was used as a backbone for a synthetic DNA library of Kozak sequences. The diversifying positions were −5, −4, −3, −2, −1, and +4 of the start codon for any of the four nucleotides. The library was synthesized by Genewiz, Inc. The library was cloned in a transient transfection vector using an ampicillin resistance marker under the transcriptional control of the CMV promoter. The library was transformed into Max Efficiency® DH5α competent cells (Invitrogen, Carlsbad, CA) and multiple dilutions were performed to optimize the selection of individual colonies. DNA preparation of individual colonies was performed according to the manufacturer's recommendations (QIAGEN, Hilden, Germany). Sequencing was performed using universal forward and universal reverse primers.

[0183] Stable vector construction

[0203] Two integrated antibody expression vectors were used to generate bispecific stable cell lines. The expression vectors contain two separate cytomegalovirus (CMV) promoters directing the transcription of both heavy chain (HC) and light chain (LC) as two separate units. The Streptomyces alboniger puromycin-N-acetyltransferase (PUR) (Vara JA, Portela A, Ortin J, Jimenez A. Nucleic Acids Res. 1986;14(11):4617-24) gene was used as a selectable marker in one plasmid. A gene encoding a fusion protein containing a positive and negative selectable marker resulting in hygromycin resistance (Hyg R) was used as a selectable marker in the other plasmid. 25 expression vectors per group were generated by cloning five heavy and five light chain combinations under different Kozak sequence variants to generate Kozak mixed pools. Two expression plasmids (one for each antibody with heavy and light chains under the Wt Kozak sequence) were used to generate the Wt Kozak pool.

[0184] Generation of stable cell lines

[0204] CHO cells were transfected using the MaxCyte STX Transfection System according to the manufacturer's recommendations (MaxCyte, Gaithersburg, MD). The transformed cells were pooled into two separate pools and selected in selective medium containing puromycin 5 μg / ml and FIAU 0.5 μM. After harvesting, one pool was subjected to single cell cloning (SCC) by limiting dilutions at 1 cell / well into 384-well clear flat-bottom tissue culture treated plates (Corning Inc, Corning, NY) using a Wellmate Microplate dispenser (Thermo Matrix) with 0.5 mm pore size Integra Viafill sterile 8-channel tubes (IntegrΛ). Plates were incubated at 37°C and 5% CO2. Three to four weeks after plating, 704 individual colonies were picked into 96-well plates (Corning Inc, Corning, NY) and assessed for antibody production using homogeneous time-resolved FRET (HTRF) approximately two days later. The top 48 followed by the top 24 antibody expressing clones were assayed via HTRF. The top 11 single cell clones were adapted to suspension growth and evaluated in production assays. Seven additional clones that showed intermediate and low HCCF titers compared to the top clones were also selected for further analysis.

[0185] Shake flask fed-batch production assay

[0205] Fed-batch production cultures were performed in shake flasks (Corning Inc, Corning, NY) containing chemically defined basal medium with bolus dosing on days 7 and 10. 6Cells were seeded at 1000 cells / ml. A temperature shift from 37°C to 35°C was performed on day 3. Titers on day 14 were determined using Protein A affinity chromatography with UV detection. Viability and viable cell counts were determined on days 0, 3, 7, 10, and 14 using a Vi-Cell XR instrument (Beckman Coulter). Glucose and lactate concentrations were monitored on days 7 and 14 using a Bioprofile 400 Analyzer (Nova Biomedical).

[0186] Antibody Surface Staining Protocol

[0206] For antibody surface staining, approximately 2 million cells were pelleted and washed twice with PBS buffer. The cells were then resuspended in 0.5 mL of PBS containing anti-human IgG (H+L) allophycocyanin (APC)-conjugated secondary antibody (Jackson Immunoresearch, West Grove, PA) at a dilution of 1:100 and incubated for 20 min at 37°C with shaking. Unstained control samples were resuspended in PBS without antibody. Empty host was also stained to set gates. After 20 min incubation with staining antibody, the cells were washed once with PBS and resuspended in 400 μl of PBS. The cells were then analyzed on a FACscan flow cytometer (Attune NxT Flow Cytometer, Life Technologies). For each analysis, 20,000 events were recorded.

[0187] Product quality analysis

[0207] Standard product quality analysis was performed by subjecting total protein A-purified antibodies to non-reducing capillary electrophoresis capillary dodecyl sodium sulfate (CE-SDS). Sample preparation was automated using a TECAN Evo200 system. All samples were diluted to 1 mg / ml to ensure consistent liquid handling and maintain optimal dye-to-protein ratios. Prepared samples were immediately analyzed on a labChip GXII system and data processed using Chromeleon software. Signal intensity, peak profile, and relative peak area distribution were assessed.

[0188] mass spectrophotometry

[0208] To identify and quantify protein product-associated variants, qualitative analysis and mass determination were performed on an Agilent 6230 Time-of-Flight mass spectrometer using HPLC-Chip separation and ion source. Data deconvolution was performed using MassHunter software. We determined the ion abundances of species with dual specificity and incorrect LC and HC stoichiometry.

[0189] Genomic DNA extraction, PCR amplification, and sequencing

[0209] Genomic DNA extraction was performed using the DNeasy Blood & Tissue Kit (QIAGEN, Hilden, Germany) according to the manufacturer's manual. Genomic DNA was quantified using a SimpliNano Microvolume Spectrophotometer (GE Lifesciences). PCR amplification was performed using Q5® Hot Start High-Fidelity 2X Master Mix (New England Biolabs). 1 μg of gDNA was used as template. PCR conditions were optimized. An annealing temperature of 64°C and 30 cycles were used. Specific annealing primers for the variable regions of each strand were designed to distinguish the same strand between molecules. The following forward primers (F) and reverse primers (R) were used to amplify Ab1 heavy chain, Ab1 light chain, Ab2 heavy chain, and Ab2 light chain: HC-Ab1: F-5'-GATACCAGCACCAGCACCGCCT-3' (SEQ ID NO: 12) and R-5'-ATGGGCGGTAGGCGTGTACGG-3' (SEQ ID NO: 13); LC-Ab1: F-5'-CTGAACAGCCGCACCCGCAA-3' (SEQ ID NO: 14) and R- HC-Ab2: 5'-GTGATTTGGCGCGGCGGCA-3' (SEQ ID NO: 16) and R-5'-ATGGGCGGTAGGCGTGTACGG-3' (SEQ ID NO: 17); and LC-Ab2: 5'-GTGCGCAACCTGGTGGTGTGG-3' (SEQ ID NO: 18) and R-5'-ATGGGCGGTAGGCGTGTACGG-3' (SEQ ID NO: 19). PCR products were cleaned using PCR purification Kit (QIAGEN) according to the manufacturer's instructions and analyzed for correct size on a 12-well, 2% pre-made agarose gel (Thermo Fisher Scientific). Sequencing was performed using the following primers: SEQ1: 5'-AACGGTGCATTGGAACGCGG-3' (SEQ ID NO: 20) and SEQ2: 5'-TGGCTTCGTTAGAACGCAGC-3' (SEQ ID NO: 21).The Kozak sequence variants of each strand of each clone were confirmed at least twice. Sequencing revealed Kozak mixed clones in which one of the strands was absent. For Ab2-LC, Ab2-HC, Ab1-LC, Ab2-HC, and Ab1-LC, these were Kozak mixed clones 3, 11, 29, 60, and 21L, respectively. Kozak mixed clone 24 does not have the HC and LC of Ab1 (Figures 23A-23C).

[0190] result

[0210] To investigate whether alterations in the consensus Kozak sequence could modulate protein production in industrialized producer cell lines, the -3 position upstream of the start codon of the ORF encoding the Fc fusion protein "A" and combinations of -3, -2, and -1 positions were varied (Figure 1). Kozak sequence variants were tested by transient transfection in CHO producer cell lines and Fc fusion protein production was assessed 48 hours post-transfection by HTRF (homogeneous time-resolved fluorescence) assay. A well-expressing and a poorly expressing antibody were also included as internal assay controls.

[0191]

[0211] As shown in Figure 2, alterations in the consensus Kozak sequence modulated the potency of Fc fusion protein A. The results confirmed that the presence of a purine at position -3 (GCCATGG and ACCATGG, SEQ ID NO: 4 and 3, respectively) produced high levels of Fc fusion protein A titer. The weakest Kozak sequence variant evaluated, TTTATGG (SEQ ID NO: 7), modulated potency at half the level of the Kozak consensus, hereafter referred to as wild-type (Wt) Kozak.

[0192]

[0212] To determine if Kozak sequence variants can modulate protein production regardless of molecule, we tested these combinations in a different Fc fusion protein: Fc fusion "B" (Figure 2). Similar to cells transfected with Kozak sequence variants of Fc fusion A, cells transfected with Kozak sequence variants of Fc fusion B showed altered titers compared to Wt Kozak, and a similar pattern was observed for the variants of Fc fusion B. These results suggested that alterations in the Wt Kozak sequence can be used to modulate antibody titers.

[0193] Example 2: Design and screening of a library of Kozak sequences to expand the range of expression levels

[0213] Additional Kozak sequence variants were tested to achieve a comprehensive range of expression levels from 0.1× to 1.0× fold over Wt Kozak. A variant library of the Kozak sequence was designed and screened using an ORF encoding Fc fusion protein B, which is representative of a normally functioning molecule. The library encompassed randomization of five bases upstream (positions −5, −4, −3, −2, and −1) and one base downstream (position +4) of the start codon to any of four nucleotides ( FIG. 4 ). The expected diversity of this construct is approximately 4100 variants of the Kozak sequence. Randomization of position +4 created an amino acid change from glycine to arginine, present in the second position of the signal sequence, when this position was occupied by adenine (A) or cytosine (C). The library was transformed into competent E. coli cells and individual variants were screened to exclude those that exhibited undesigned mutations, deletions, and variants containing a stop codon as the second amino acid of the signal sequence, which occurred when position +4 was occupied by thymine (T).

[0194] result

[0214] An initial screening by transient transfection in CHO cells was performed with 111 correct variants, including Wt Kozak and Wt Kozak with C or A at position +4 as controls for the change of the second amino acid of the signal sequence. As shown in Figure 5, Fc fusion protein titers measured 48 hours after transient transfection by HTRF assay showed a wide range of expression levels. A nearly continuous range of translation was observed with normalized titers of 0.05 to 1.54 units, where 1.0 corresponds to Wt Kozak. These variants met the desired expression range target such that further variant screening was stopped. In general, the presence of A or C at position +4 increased the titer over Wt Kozak, regardless of the Kozak sequence upstream of the start codon. The screening was repeated and data from two independent transient transfections are presented (Figure 5).

[0195]

[0215] As shown in Figure 6, the nucleotide distribution of the analyzed variants was randomized, indicating that the library screening was not skewed. The variants were then classified into five groups based on their respective titers and performance. The ranges of the groups were as follows: Group I, 0.05-0.3; Group II, 0.31-0.6; Group III, 0.61-0.8; Group IV, 0.81-1.00; and Group V, 1.01-1.34. No other correlations between antibody production and Kozak sequence variants were detected, except for the preference for adenine and thymine at most positions in the weakest group, and arginine as the second amino acid of the signal sequence in the strongest group. To reduce the number of variants, transient transfections were performed several times (Figure 7). The criteria for round-to-round selection of variants were reproducibility between transient transfections, stability between different DNA preparations, and conservation of nucleotide diversity. As shown in Figures 8A-8B, this screening established 11 Kozak sequence variants. This represents an expression range of 0.2 to 1.3-fold that of Wt Kozak, allowing for titer control with high precision.

[0196] Example 3: Generation of bispecific stable clones under a mixture of Kozak sequence variants

[0216] Although progress has been made in engineering complex antibody formats such as bispecific antibodies, manufacturability in a single mammalian expression system often shows poor performance. Low titers and poor product quality are two of the factors that make stable production of bispecific antibodies (BsAbs) difficult. As a result, there is an urgent need to identify the limiting steps of the production system. One bottleneck that limits the efficiency of assembly of bispecific and other multichain formats in single cells is the limited control at the individual chain level. Designing vectors that allow tuning of the ratio of individual chains in multichain formats can improve assembly efficiency as previously demonstrated in E. coli using translation initiation region (TIR) ​​variants (Simmons LC, Yansura DG. Nat Biotechnol. 1996;14(5):629-34. For example, as shown in Figure 9, the ratio of light chains that make up the BsAb (LC1:LC2) can be manipulated to result in a higher percentage of correctly assembled BsAbs.

[0197]

[0217] To generate bispecific antibodies, it is necessary to identify unique parental antibody ratios for merging into a bispecific format, and therefore the ability of Kozak sequence variants to result in higher assembly and production of BsAbs in single cells was tested. The successful application of Kozak sequence variants as a technique to modulate protein expression was evaluated in the establishment of stable host cell lines. To do so, a representative panel of five variants including Wt Kozak was selected (Figure 10 and Figure 8A). The relative protein production intensities of these variants were approximately 0.3x, 0.5x, 0.8x, 1.0x, and 1.3x fold higher than that of Wt Kozak sequence for Kozak sequence variants #3, #135, #148, Wt, and #228, respectively. Thus, the panel of variants covered the wide range of expression levels observed in transient transfections (Figure 10).

[0198]

[0218] The bispecific antibody Ab1 / Ab2 was chosen as a model to evaluate the effect of the Kozak sequence on tailoring the chain ratio based on the following criteria: First, a well-characterized process development platform was available for Ab1 / Ab2 that facilitated product analysis and characterization. Second, Ab1 is a poorly expressed molecule with a bottleneck downstream of the transcription process, whereas Ab2 is an example of a well-functioning molecule. This is a suitable scenario to test the ability of Kozak sequence variants to create successful chain pairing.

[0199]

[0219] To preferentially promote heavy chain heterodimerization, the heavy chains (HC) of both Ab2 and Ab1 carried knob and hole mutations, respectively. Point mutations were used to mitigate light chain (LC) mispairing (Dillon M, Yin Y, Zhou J, McCarty L, Ellerman D, Slaga D, et al. MAbs. 2017;9(2):213-30). Twenty-five integrated expression vectors were generated using combinations of heavy and light chains under the previously described panel of five Kozak sequence variants (Figure 11). A mixture of 50 plasmids was transfected into CHO cells (hereafter referred to as Kozak mix). The diversity generated with this approach reached a potential of 625 different chain ratio combinations. As a baseline, four chains were cloned of BsAb under the Wt Kozak sequence, transfected into CHO cells, and carried in parallel (hereafter referred to as Wt Kozak). Two different transfections were performed for each condition.

[0200] result

[0220] Transfection efficiency was monitored by cell surface staining for total IgG and measured by FACS analysis. Figure 12 shows that similar transfection efficiency was observed for both Kozak mixed pools and Wt Kozak pools. After transfection of expression plasmids, stable cell pools were established by drug selection. As shown in Figures 13A and 13B, similar cell viability was observed for both Kozak mixed pools and Wt Kozak pools after addition of selection drug (indicated by black arrows). One stable pool from each transfection in each condition was then selected for single cell cloning. After recovery, plates were analyzed for clone recovery rate and 704 recovered clones per pool and condition were picked into 96-well plates. Primary screening based on harvested cell culture fluid (HCCF) titers showed differences in profile and absolute titers between conditions. Kozak mixed clones showed lower absolute titers and a more pronounced decrease in titer than Wt Kozak clones (Figures 14A and 14B). This result is consistent with the diversity of chain ratio combinations planned in the Kozak mixed clones, while the profile of the Wt Kozak clones was as expected for clones carrying only one chain ratio combination. Several rounds of screening based on HCCF titer, suspension adaptation, and scale-up were completed to determine the top 11 clones for each transfection pool and each condition, generating a total of 22 clones for each condition. Due to the fact that each Kozak mixed clone has the potential to carry different Kozak sequence variant combinations, and to understand the differences between the combinations, an additional seven clones that showed moderate and low HCCF titers compared to the top clones were also selected for further analysis.

[0201] Example 4: Shake-flash performance of bispecific stable clones derived from a mix of Kozak sequence variants

[0221] Shake flask fed-batch production for 14 days was used to evaluate the productivity and bispecific assembly of each individual clone selected from Example 3 above. All clones investigated showed comparable viability throughout the production assay. Final viability on day 14 was approximately 80-95% in both conditions, except for one Kozak mixed clone which resulted in 74% (Figures 15A & 15B). Similarly, viable cell counts for all individual clones showed exponential growth rates until day 7, after which they reached a plateau phase (Figures 16A and 16B). Product quality and assembly efficiency were measured on day 14 for overall titers of each clone.

[0202]

[0222] As shown in Figure 17, individual Wt Kozak clones showed higher general absolute antibody titers than individual Kozak mixed clones in both transfections. It should be noted that correctly assembled bispecific antibodies, half antibodies, and other undesired by-products are all represented in this titer. To assess product quality, samples were analyzed by non-reducing capillary electrophoresis sodium dodecyl sulfate (CE-SDS) to provide information on molecular weight forms and other impurities. This technique allowed the differentiation and quantification of the main peak equated to full antibody (full-ab) in two formats: correctly assembled and incorrectly assembled BsAb formats, and the sum of pre-peaks representing other species of different molecular weight. Overall, it was observed that fewer Kozak mixed clones had a higher percentage of full-ab formation compared to Wt Kozak clones among the top 22 clones (Figure 18). This result is consistent with the fact that although the chain ratio combinations differed between conditions (1 vs. 625 for Wt Kozak clones and Kozak mixed clones, respectively), a similar number of individual clones were screened and selected for each condition, which favored finding more Wt Kozak clones with high product quality. The percentage of complete antibody varied from clone to clone. The four individual Kozak mixed clones from transfection 2 ranged around 60% complete antibody, then dropped to 30%.

[0203]

[0223] All of the individual Kozak mixed clones from transfection 1 showed approximately 30% of the main peak (Figure 18). This result is consistent with the fact that in the diversity of the chain ratio combinations in the Kozak mixed pool, some of them were weak in terms of producing full ab. In addition, not all of the highest antibody producing clones (overall titers) in the Kozak mixed clones were associated with the highest content of full ab.

[0204]

[0224] In general, the best single Wt Kozak clones ranged from 80-90% of full ab production. Similar to the Kozak mixed clones, individual Wt Kozak clones showed variability in the percentage of the main peak depending on the clone. We also observed a correlation between the high overall titers of these clones and the high content of the main peak (Figure 18).

[0205]

[0225] In parallel, seven additional clones with moderate and low HCCF titers compared to the top clones were evaluated for their production performance. The production titers of these clones showed similar behavior as previously observed in the primary screen (Figure 19). CE-SDS data showed that the product quality varied from clone to clone. Again, there was no direct correlation between the general titer and the highest content of complete antibodies. In this manner, some Kozak mixed clones with more moderate titers compared to the top 11 clones showed a similar percentage of complete ab, around 50-60% (Figure 19B).

[0206] Example 5: Kozak sequence variants enhanced bispecific assembly in stable clones

[0226] The top four clones from the CE-SDS analysis per condition and transfection were analyzed to identify and quantify HC and LC mispairings. IgG was recovered from cell culture media using protein A affinity chromatography, and the resulting pool was analyzed by intact liquid chromatography mass spectrometry (LC-MS). After data deconvolution, multiple product-associated variants were identified, including half antibodies (HC+LC), homodimers (two of the same HC), and mispaired LC-HC species. Using the calculations reported in the Methods section, the LC-MS data was used to orthogonally weight the CE-SDS results, reporting the proportion of species that analytically overlap in CE-SDS due to having similar masses.

[0207] result

[0227] Seven different bispecific antibody formats and three half antibody formats were differentiated and quantified (Figure 20A). The best individual Kozak mixed clone showed approximately 40% of correctly assembled BsAb, more than double that observed for the best individual Wt Kozak clone (approximately 18%) (Figure 20A). Because each individual Kozak mixed clone potentially harbors a different combination of Kozak sequence variants, all clones showed different percentages of BsAb assembly, whereas Wt Kozak clones harboring the same combinations showed similar amounts of BsAb assembly. Of note, the incorrectly assembled bispecific formats Ab1 -HC + Ab2 -HC + 2x Ab2 -LC (which had light chain mispairings) ranged from 41% to 64% of the complete antibody produced by the Wt Kozak clone (Figure 20A). These data point out the importance of adjusting the light chain ratio to achieve BsAb assembly when using single cell production. Mass spectrometry analysis of the Kozak mixed clones with the lowest complete ab formation yields as measured by CE-SDS confirmed that the majority of the product of those clones was half ab. The formats were Ab1-HC + Ab2-LC and whole half Ab1. The BsAb formats exhibited by these clones were mostly incorrectly assembled BsAb 2x Ab1-HC + 2x Ab2-LC, with a small proportion in the BsAb format 2xAb2-HC + Ab1-LC + Ab2-LC (Figure 20A).

[0208]

[0228] Understanding the percentages of each format, the effective bispecific titers were calculated (Figure 20B). The Kozak mixed clones overcome the general titer deficiencies shown earlier due to the fact that some Kozak mixed clones have higher BsAb assembly than Wt Kozak clones (Figure 17 and Figures 20A-20B). The effective bispecific titers were similar between the best clones for each condition at 0.7g / L and 0.6g / L for the best Wt Kozak and Kozak mixed clones, respectively. As every Kozak mixed clone potentially harbors a different combination of Kozak variants, each clone showed a different percentage of bsAb assembly, whereas WT Kozak clones harboring the same combination showed a much narrower range of bsAb assembly (Figure 20A). Of note, the incorrectly assembled bispecific formats Ab2 HC + Ab1 HC + 2x Ab2 LC (which had light chain mispairing) ranged from 41-64% of the complete ab produced by the Wt Kozak clone (Figure 20A). These data point out the importance of tuning the light chain translation levels to achieve correct bsAb assembly when using single cell production.

[0209]

[0229] However, as shown in Figure 20B, Kozak clones show lower amounts of impurities to be removed during purification compared to Wt Kozak clones. The fact that some Kozak mixed clones have higher BsAb assembly than Wt Kozak clones overcomes the general titer deficiency shown earlier in Figure 17. Importantly, Kozak mixed clones also showed lower amounts of product-associated impurities than WT Kozak clones (Figure 20B). Purification of correctly assembled bsAbs from numerous nearly identical potential by-products is challenging. Some of the by-products are so similar to the target bsAb that their removal downstream is often done at the expense of yield or product quality. New analytical methods were needed to differentiate these inherent impurities from the desired product, but they require high manufacturing costs and complex technologies. Kozak mixed clones minimized impurities and product-associated variant formation (Figures 20B and 22B), demonstrating the benefits of this approach to produce bsAbs in single cells.

[0210]

[0230] These data indicate that tuning translation strength using Kozak sequence variants improves assembly of bispecific antibodies and reduces product-associated impurities in CHO cells.

[0211] Example 6: High yields of BsAb are produced by adjusting the chain ratio with Kozak sequence variants

[0231] The heterogeneity and diversity observed among the Kozak mixed clones in terms of BsAb assembly efficiency and product quality suggested that each clone had chain expression under different Kozak sequence variants. Sequencing of 29 Kozak mixed clones was performed to identify which Kozak combination each clone had the best combination and to determine the Kozak combination.

[0212] result

[0232] The data showed that Kozak mixed clones with low product yields had Ab2 heavy chain expression under either the weakest Kozak sequence variants Kz.135 and Kz.3, with the remainder of the chain having either Wt or the strongest Kozak sequence variant (Figures 21A-21B). As a result, insufficient amounts of Ab2 heavy chain were likely produced, and consequently the final product was preferentially half ab in the format Ab1 -HC + Ab2 -LC, or whole half Ab2 (Figure 20A). In other words, there was a range in which heavy chain expression could be reduced before compromising full ab assembly.

[0213]

[0233] Alternatively, when the combination of both Wt Kozak and the strongest Kozak sequence variant regulated the expression of four chains, Kozak mixed clones showed high BsAb assembly with high titers (Figure 21A). On the other hand, when the combination of both WT Kozak and the strongest Kozak sequence variant (Kz.228; relative intensity of 1.3) regulated the expression of four chains, Kozak mixed clones showed high bsAb assembly and effective titers (Figures 20A and 21A, Kozak mixed clones 69, 61, and 88). Interestingly, a slight light chain ratio of 1.3x fold of Ab1 light chain to 1.0x fold of Ab2 light chain, or vice versa, as shown by Kozak mixed clone 69 and Kozak mixed clone 61, respectively, significantly reduced the Ab2 light chain mispairing observed in WT Kozak clones (Figure 20A; Ab2 HC + Ab1 HC + 2x Ab2 LC). These data suggested that reduced translation of the selected chain could efficiently increase bsAb assembly relative to the translation level of the other chain and improve the overall product quality by limiting the accumulation of variants. However, expression of any of the light chains under the strongest Kozak sequence variant promoted the production of more by-products composed by that chain instead of the more correctly assembled BsAb. For example, Kozak clone 69, which has light chain Ab1 under the strongest Kozak sequence, showed about 30% of the product as Whole 1 / 2 Ab1 (Figure 20A). This observation was consistent with the fact that the production of BsAbs requires stoichiometric chain pairing to merge into a bispecific format. In this regard, saturation of the protein production system may result in the accumulation of variants.

[0214]

[0234] Given this, and not wishing to be bound by theory, it is believed that reduced translation can prevent the accumulation of variants and result in increased yields. Consistent with this conclusion, one combination performed at half the titer of the top clone and carrying both the heavy and light chains of Ab1 under a weak Kozak sequence, variant #135, showed the best bispecific combination detected (Figure 22D). CE-SDS data for this clone showed that the production quality was about 48% of the full ab and about 52% of the half ab. MS data showed that 100% of the full ab corresponded only to the bispecific format. This is about 50% of the correctly assembled BsAbs. Meanwhile, nearly all of the half ab species were knob 1 / 2 Ab2 (Figures 22A-22B). This is consistent with the fact that the heavy and light chains of Ab2 were under the Wt Kozak sequence and the strongest Kozak sequence variant #228, respectively. Thus, within certain limits, downregulation of expression of both chains of molecules that are likely to represent a bottleneck downstream of translation during the production process had a positive impact on assembly efficiency.

[0215]

[0235] Clone 17M was one of the additional clones that was carried forward, although the primary HCCF titers were moderate. This clone had only roughly half the effective titers of the top Kozak clones, but showed the best bispecific assembly at approximately 48%. MS data showed that 100% of the full ab corresponded only to the correct bispecific format, with the half ab species being almost exclusively restricted to knob1 / 2 mAb2 (Figures 22A and 22B).

[0216]

[0236] These data can be explained by the results of sequencing (Figure 22C). Both the heavy and light chains of Ab1 were under a weak Kozak variant (Kz.135), but as previously described, Ab1 is a difficult to express molecule. Downregulation of both chains could alleviate downstream bottlenecks of translation, such as protein folding, when using Kz.135, which has a relative strength of approximately 0.5, and had a positive effect on assembly efficiency. For Ab2, the heavy and light chains were under the stronger Kozak sequence, WT Kozak and the strongest Kozak variant (Kz.228), respectively (Figure 22C).

[0217]

[0237] This combination, together with the weaker Kozak sequence of aAb1, results in accumulation of the Ab2 half-ab while limiting significant accumulation of other species.

[0218]

[0238] In the same manner, we sequenced seven additional clones with intermediate and low HCCF titers compared to the top clones. In general, we observed that the expression of most chains of both molecules was controlled by the weakest Kozak sequence variants among these clones (Figure 21C). This result confirms the poor performance of these clones compared to the top clones.

[0219]

[0239] To elucidate Ab species and BsAb formats, five of these Kozak mixed clones, which harbored combinations of Kozak sequence variants significantly different from the top 11 clones, were analyzed by mass spectrometry. With the exception of Kozak mixed clone 17, the percentage of BsAb assemblies was less than 10%. The absolute abundance of each species correlated with the expression intensity of each chain (Figure 22D).

[0220]

[0240] The foregoing disclosure has been described in some detail by way of illustration and example for purposes of clarity of understanding, but the descriptions and examples should not be construed as limiting the scope of the disclosure. The disclosures of all patent and scientific literature cited herein are expressly incorporated by reference in their entirety.

Claims

1. 1. A method for producing a bispecific antibody in a eukaryotic host cell, the bispecific antibody comprising a first half antibody comprising a first antibody heavy chain and a first antibody light chain, and a second half antibody comprising a second antibody heavy chain and a second antibody light chain; below: (a) providing a eukaryotic host cell, a eukaryotic host cell comprising a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding a first antibody heavy chain, a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding a first antibody light chain, a third polynucleotide comprising a third translation initiation sequence operably linked to a third open reading frame encoding a second antibody heavy chain, and a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to a fourth open reading frame encoding a second antibody light chain; when each half antibody is expressed individually in a eukaryotic host cell, the first half antibody is expressed at a lower level than the second half antibody; one or both of the first translation initiation sequence and the second translation initiation sequence are weaker than one or both of the third translation initiation sequence and the fourth translation initiation sequence; Providing a eukaryotic host cell; (b) culturing the eukaryotic host cell under conditions suitable for expression of the first and second antibody heavy and light chains, wherein the first and second antibody heavy and light chains form a bispecific antibody, wherein the first half antibody binds to a first antigen and the second half antibody binds to a second antigen; (c) recovering the bispecific antibody produced by the eukaryotic host cell; and A method comprising:

2. 2. The method of claim 1, wherein one or both of the first antibody heavy chain and the first antibody light chain of the first half antibody is expressed at a lower level than one or both of the second antibody heavy chain and the second antibody light chain of the second half antibody when both half antibodies are expressed in the same host cell.

3. 3. The method of claim 1 or 2, wherein the first, second, third, and fourth translation initiation sequences all comprise the sequence (5' to 3') NNNNNATGNGA, where N is C, G, A, or T / U (SEQ ID NO:1).

4. 4. The method of any one of claims 1 to 3, wherein one or both of the first translation initiation sequence and the second translation initiation sequence comprises a sequence selected from the group consisting of SEQ ID NOs: 8-10.

5. 5. The method of claim 1, wherein one or both of the third translation initiation sequence and the fourth translation initiation sequence comprises the sequence ACCATGG (SEQ ID NO: 3) or GAAGTATGA (SEQ ID NO: 11).

6. 3. The method of claim 1 or 2, wherein the first translation initiation sequence comprises the sequence of SEQ ID NO:9, the second translation initiation sequence comprises the sequence of SEQ ID NO:9, the third translation initiation sequence comprises the sequence of SEQ ID NO:2, and the fourth translation initiation sequence comprises the sequence of SEQ ID NO:

11.

7. 7. The method of any one of claims 1 to 6, wherein each of the first, second, third, and fourth polynucleotides is operably linked to a promoter.

8. 8. The method of claim 7, wherein the first polynucleotide and the second polynucleotide are operably linked to the same promoter, and the third polynucleotide and the fourth polynucleotide are operably linked to the same promoter.

9. 9. The method of claim 1, wherein the first translation initiation sequence is weaker than the third translation initiation sequence.

10. 10. The method of claim 1, wherein the second translation initiation sequence is weaker than the fourth translation initiation sequence.

11. 11. The method of claim 1, wherein the first translation initiation sequence is weaker than the fourth translation initiation sequence.

12. 12. The method of claim 1, wherein the second translation initiation sequence is weaker than the third translation initiation sequence.

13. 13. The method of any one of claims 1 to 12, wherein the first translation initiation sequence is the same as the second translation initiation sequence.

14. 14. The method of any one of claims 1 to 13, wherein the third translation initiation sequence is the same as the fourth translation initiation sequence.

15. 2. The method of claim 1, wherein the first antibody heavy chain comprises a first antibody Fc region comprising a CH2 domain and a CH3 domain; and the second antibody heavy chain comprises a second antibody Fc region comprising a CH2 domain and a CH3 domain; and the CH3 domain of the first antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a smaller side chain volume, thereby generating a hole on the surface of the CH3 domain of the first antibody Fc region that interacts with the CH3 domain of the second antibody Fc region; and the CH3 domain of the second antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a larger side chain volume, thereby generating a knob on the surface of the CH3 domain of the second antibody Fc region that interacts with the CH3 domain of the first antibody Fc region.

16. 2. The method of claim 1, wherein the first antibody heavy chain comprises a first antibody Fc region comprising a CH2 domain and a CH3 domain; and the second antibody heavy chain comprises a second antibody Fc region comprising a CH2 domain and a CH3 domain; and the CH3 domain of the second antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a smaller side chain volume, thereby generating a hole on the surface of the CH3 domain of the second antibody Fc region that interacts with the CH3 domain of the first antibody Fc region; and the CH3 domain of the first antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a larger side chain volume, thereby generating a knob on the surface of the CH3 domain of the first antibody Fc region that interacts with the CH3 domain of the second antibody Fc region.

17. 17. The method of claim 15 or 16, wherein the knob mutations include at least one of T366Y, T366W, T394W, and F405W, numbered based on human IgG1 according to the EU index.

18. 18. The method of any one of claims 15 to 17, wherein the hole mutations include at least one of F405A, Y407T, Y407A, T366S, L368A, Y407V, and T394S, numbered based on human IgG1 according to the EU index.

19. 19. The method of any one of claims 15 to 18, wherein the knob mutation comprises T366W, and wherein the hole mutation comprises at least one, at least two, or all three of T366S, L368A, and Y407V, numbered based on human IgG1 according to the EU index.

20. 20. The method of any one of claims 1 and 15 to 19, wherein the first antibody light chain comprises a first mutation and the first antibody heavy chain comprises a second mutation, and the first and second mutations promote selective association of the first antibody light chain with the first antibody heavy chain.

21. 21. The method of claim 20, wherein the first mutation comprises an amino acid substitution at V133 and the second mutation comprises an amino acid substitution at S183 (numbering according to the EU index).

22. 22. The method of claim 21 , wherein the S183 substitution is selected from the group consisting of S183A, S183T, S183V, S183Y, S183F, S183H, S183N, S183D, S183E, S183R, and S183K, and the V133 substitution is selected from the group consisting of V133E, V133S, V133L, V133W, V133K, V133R, and V133D.

23. 23. The method of claim 21 or 22, wherein the amino acid substitution at S183 results in a positively charged residue and the amino acid substitution at V133 results in a negatively charged residue, or wherein the amino acid substitution at S183 results in a negatively charged residue and the amino acid substitution at V133 results in a positively charged residue.

24. 20. The method of any one of claims 1 and 15 to 19, wherein the second antibody light chain comprises a third mutation and the second antibody heavy chain comprises a fourth mutation, the third and fourth mutations promoting selective association of the second antibody light chain with the second antibody heavy chain.

25. 25. The method of claim 24, wherein the third mutation comprises an amino acid substitution at V133 and the fourth mutation comprises an amino acid substitution at S183 (numbering according to the EU index).

26. 26. The method of claim 25, wherein the S183 substitution is selected from the group consisting of S183A, S183T, S183V, S183Y, S183F, S183H, S183N, S183D, S183E, S183R, and S183K, and the V133 substitution is selected from the group consisting of V133E, V133S, V133L, V133W, V133K, V133R, and V133D.

27. 27. The method of claim 25 or 26, wherein the amino acid substitution at S183 results in a positively charged residue and the amino acid substitution at V133 results in a negatively charged residue, or wherein the amino acid substitution at S183 results in a negatively charged residue and the amino acid substitution at V133 results in a positively charged residue.

28. 20. The method of any one of claims 1 and 15 to 19, wherein a first antibody light chain comprises a V133K mutation, a first antibody heavy chain comprises a S183E mutation, a second antibody light chain comprises a V133E mutation, and a second antibody heavy chain comprises a S183K mutation (numbering according to the EU index).

29. 29. The method of claim 28, wherein the first antibody heavy chain further comprises a T366S, L368A, and Y407V mutation and the second antibody heavy chain further comprises a T366W mutation (numbered based on human IgG1 according to the EU index).

30. 20. The method of any one of claims 1 and 15 to 19, wherein the second antibody light chain comprises a V133K mutation, the second antibody heavy chain comprises a S183E mutation, the first antibody light chain comprises a V133E mutation, and the first antibody heavy chain comprises a S183K mutation (numbering according to the EU index).

31. 31. The method of claim 30, wherein the second antibody heavy chain further comprises a T366S, L368A, and Y407V mutation and the first antibody heavy chain further comprises a T366W mutation (numbered based on human IgG1 according to the EU index).

32. 32. The method of any one of claims 1 to 31, wherein the first, second, third and fourth polynucleotides are integrated into one or more chromosomes of the eukaryotic host cell.

33. 33. The method of claim 32, wherein the first, second, third, and fourth polynucleotides are integrated into the same chromosomal locus of the eukaryotic host cell.

34. 32. The method of any one of claims 1 to 31, wherein the first, second, third and fourth polynucleotides are part of one or more extrachromosomal polynucleotides in a eukaryotic host cell.

35. 35. The method of any one of claims 1 to 34, wherein the eukaryotic host cell is a mammalian host cell.

36. 36. The method of claim 35, wherein the mammalian host cell is a Chinese Hamster Ovary (CHO) cell.

37. 37. The method of any one of claims 1 to 36, wherein the first and / or second translation initiation sequence is at least 5%, at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, or at least 100% weaker than the third and / or fourth translation initiation sequence.

38. 37. The method of any one of claims 1 to 36, wherein the first half antibody is expressed at a level that is at least 5%, at least 10%, at least 15%, at least 20%, at least 25%, at least 30%, at least 35%, at least 40%, at least 45%, at least 50%, at least 55%, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, or at least 100% lower than the expression level of the second half antibody when each half antibody is expressed individually in a eukaryotic host cell.

39. 40. A method for producing a composition comprising one or more bispecific antibodies, the method comprising producing each bispecific antibody of the one or more bispecific antibodies by a method according to any one of claims 1 to 38.

40. 1. A recombinant eukaryotic host cell for the expression of a bispecific antibody comprising a first half antibody comprising a first antibody heavy chain and a first antibody light chain, and a second half antibody comprising a second antibody heavy chain and a second antibody light chain, (a) a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding a first antibody heavy chain; (b) a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding the first antibody light chain; (c) a third polynucleotide comprising a third translation initiation sequence operably linked to a third open reading frame encoding a second antibody heavy chain; (d) a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to a fourth open reading frame encoding a second antibody light chain; Including, when each half antibody is expressed individually in a recombinant eukaryotic host cell, the first half antibody is expressed at a lower level than the second half antibody; one or both of the first translation initiation sequence and the second translation initiation sequence are weaker than one or both of the third translation initiation sequence and the fourth translation initiation sequence; Recombinant eukaryotic host cells.

41. 41. The cell of claim 40, wherein one or both of the first antibody heavy chain and the first antibody light chain of the first half antibody is expressed at a lower level than one or both of the second antibody heavy chain and the second antibody light chain of the second half antibody when both half antibodies are expressed in the same host cell.

42. 42. The cell of claim 40 or 41, wherein the first, second, third, and fourth translation initiation sequences all comprise the sequence (5' to 3') NNNNNATGNGA, where N is C, G, A, or T / U (SEQ ID NO: 1).

43. 43. The cell of any one of claims 40 to 42, wherein one or both of the first translation initiation sequence and the second translation initiation sequence comprises a sequence selected from the group consisting of SEQ ID NOs: 8-10.

44. 44. The cell of any one of claims 40 to 43, wherein one or both of the third translation initiation sequence and the fourth translation initiation sequence comprises the sequence ACCATGG (SEQ ID NO: 3) or GAAGTATGA (SEQ ID NO: 11).

45. 43. A cell according to any one of claims 40 to 42, wherein the first translation initiation sequence comprises the sequence of SEQ ID NO:9, the second translation initiation sequence comprises the sequence of SEQ ID NO:9, the third translation initiation sequence comprises the sequence of SEQ ID NO:2, and the fourth translation initiation sequence comprises the sequence of SEQ ID NO:

11.

46. 46. ​​The cell of any one of claims 40 to 45, wherein each of the first, second, third, and fourth polynucleotides is operably linked to a promoter.

47. 47. The cell of claim 46, wherein the first polynucleotide and the second polynucleotide are operably linked to the same promoter, and the third polynucleotide and the fourth polynucleotide are operably linked to the same promoter.

48. 48. The cell of any one of claims 40 to 47, wherein the first translation initiation sequence is weaker than the third translation initiation sequence.

49. 49. The cell of any one of claims 40 to 48, wherein the second translation initiation sequence is weaker than the fourth translation initiation sequence.

50. 50. The cell of any one of claims 40 to 49, wherein the first translation initiation sequence is weaker than the fourth translation initiation sequence.

51. 51. The cell of any one of claims 40 to 50, wherein the second translation initiation sequence is weaker than the third translation initiation sequence.

52. 52. The cell of any one of claims 40 to 51, wherein the first translation initiation sequence is the same as the second translation initiation sequence.

53. 53. The cell of any one of claims 40 to 52, wherein the third translation initiation sequence is the same as the fourth translation initiation sequence.

54. 41. The cell of claim 40, wherein the first antibody heavy chain comprises a first antibody Fc region comprising a CH2 domain and a CH3 domain; and the second antibody heavy chain comprises a second antibody Fc region comprising a CH2 domain and a CH3 domain; and the CH3 domain of the first antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a smaller side chain volume, thereby generating a hole on the surface of the CH3 domain of the first antibody Fc region that interacts with the CH3 domain of the second antibody Fc region; and the CH3 domain of the second antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a larger side chain volume, thereby generating a knob on the surface of the CH3 domain of the second antibody Fc region that interacts with the CH3 domain of the first antibody Fc region.

55. 41. The cell of claim 40, wherein the first antibody heavy chain comprises a first antibody Fc region comprising a CH2 domain and a CH3 domain; and the second antibody heavy chain comprises a second antibody Fc region comprising a CH2 domain and a CH3 domain; and the CH3 domain of the second antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a smaller side chain volume, thereby generating a hole on the surface of the CH3 domain of the second antibody Fc region that interacts with the CH3 domain of the first antibody Fc region; and the CH3 domain of the first antibody Fc region is modified such that one or more amino acid residues within the CH3 / CH3 interface are replaced with amino acid residues having a larger side chain volume, thereby generating a knob on the surface of the CH3 domain of the first antibody Fc region that interacts with the CH3 domain of the second antibody Fc region.

56. 56. The cell of claim 54 or 55, wherein the knob mutations include at least one of T366Y, T366W, T394W, and F405W, numbered based on human IgG1 according to the EU index.

57. 57. The cell of any one of claims 54 to 56, wherein the whole mutations include at least one of F405A, Y407T, Y407A, T366S, L368A, Y407V, and T394S, numbered based on human IgG1 according to the EU index.

58. 58. The cell of any one of claims 54 to 57, wherein the knob mutation comprises T366W, and wherein the hole mutation comprises at least one, at least two, or all three of T366S, L368A, and Y407V, numbered based on human IgG1 according to the EU index.

59. 59. The cell of any one of claims 40 and 54-58, wherein the first antibody light chain comprises a first mutation and the first antibody heavy chain comprises a second mutation, and the first and second mutations promote selective association of the first antibody light chain with the first antibody heavy chain.

60. 60. The cell of claim 59, wherein the first mutation comprises an amino acid substitution at V133 and the second mutation comprises an amino acid substitution at S183 (numbering according to the EU index).

61. 61. The cell of claim 60, wherein the S183 substitution is selected from the group consisting of S183A, S183T, S183V, S183Y, S183F, S183H, S183N, S183D, S183E, S183R, and S183K, and the V133 substitution is selected from the group consisting of V133E, V133S, V133L, V133W, V133K, V133R, and V133D.

62. 62. The cell of claim 60 or 61, wherein the amino acid substitution at S183 results in a positively charged residue and the amino acid substitution at V133 results in a negatively charged residue, or wherein the amino acid substitution at S183 results in a negatively charged residue and the amino acid substitution at V133 results in a positively charged residue.

63. 59. The cell of any one of claims 40 and 54-58, wherein the second antibody light chain comprises a third mutation and the second antibody heavy chain comprises a fourth mutation, the third and fourth mutations promoting selective association of the second antibody light chain with the second antibody heavy chain.

64. 64. The cell of claim 63, wherein the third mutation comprises an amino acid substitution at V133 and the fourth mutation comprises an amino acid substitution at S183 (numbering according to the EU index).

65. 65. The cell of claim 64, wherein the S183 substitution is selected from the group consisting of S183A, S183T, S183V, S183Y, S183F, S183H, S183N, S183D, S183E, S183R, and S183K, and the V133 substitution is selected from the group consisting of V133E, V133S, V133L, V133W, V133K, V133R, and V133D.

66. 66. The cell of claim 64 or 65, wherein the amino acid substitution at S183 results in a positively charged residue and the amino acid substitution at V133 results in a negatively charged residue, or wherein the amino acid substitution at S183 results in a negatively charged residue and the amino acid substitution at V133 results in a positively charged residue.

67. 59. The cell of any one of claims 40 and 54 to 58, wherein a first antibody light chain comprises a V133K mutation, a first antibody heavy chain comprises a S183E mutation, a second antibody light chain comprises a V133E mutation, and a second antibody heavy chain comprises a S183K mutation (numbering according to the EU index).

68. 68. The cell of claim 67, wherein the first antibody heavy chain further comprises a T366S, L368A, and Y407V mutation and the second antibody heavy chain further comprises a T366W mutation (numbered based on human IgG1 according to the EU index).

69. 59. The cell of any one of claims 40 and 54 to 58, wherein the second antibody light chain comprises a V133K mutation, the second antibody heavy chain comprises a S183E mutation, the first antibody light chain comprises a V133E mutation, and the first antibody heavy chain comprises a S183K mutation (numbered according to the EU index).

70. 70. The cell of claim 69, wherein the second antibody heavy chain further comprises a T366S, L368A, and Y407V mutation and the first antibody heavy chain further comprises a T366W mutation (numbered based on human IgG1 according to the EU index).

71. 71. The cell of any one of claims 40 to 70, wherein the first, second, third, and fourth polynucleotides are integrated into one or more chromosomes of the eukaryotic host cell.

72. 72. The cell of claim 71, wherein the first, second, third, and fourth polynucleotides are integrated into the same chromosomal locus of the eukaryotic host cell.

73. 71. The cell of any one of claims 40 to 70, wherein the first, second, third, and fourth polynucleotides are part of one or more extrachromosomal polynucleotides in a eukaryotic host cell.

74. 74. The cell of any one of claims 40 to 73, wherein the eukaryotic host cell is a mammalian host cell.

75. The cell of claim 74, wherein the bispecific antibody is highly produced in a eukaryotic host cell.

76. 1. A method for identifying a combination of translation initiation sequences for expressing a bispecific antibody in a eukaryotic host cell, the bispecific antibody comprising a first half antibody comprising a first antibody heavy chain and a first antibody light chain, and a second half antibody comprising a second antibody heavy chain and a second antibody light chain; (a) providing a library comprising a plurality of eukaryotic host cells, wherein each eukaryotic host cell of the plurality of eukaryotic host cells comprises: a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding a first antibody heavy chain; a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding the first antibody light chain; a third polynucleotide comprising a third translation initiation sequence operably linked to a third open reading frame encoding a second antibody heavy chain; a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to a fourth open reading frame encoding a second antibody light chain; Including, wherein multiple combinations of the first, second, third, and fourth translation initiation sequences are expressed in multiple eukaryotic host cells. Preparing a library, (b) culturing the library of eukaryotic host cells under conditions suitable for expression of the bispecific antibody by the plurality of eukaryotic host cells; (c) measuring the amount of bispecific antibody expressed by a single eukaryotic host cell of the plurality of eukaryotic host cells or a clone of a single eukaryotic host cell of the plurality of eukaryotic host cells; (d) identifying the first, second, third, and fourth translation initiation sequences of one or more of a single eukaryotic host cell of the plurality of eukaryotic host cells or a clone of a single eukaryotic host cell of the plurality of eukaryotic host cells that expresses the bispecific antibody; Including, when each half antibody is expressed individually in a eukaryotic host cell, the first half antibody is expressed at a lower level than the second half antibody, the first half antibody binds to a first antigen, and the second half antibody binds to a second antigen; method.

77. 77. The method of claim 76, wherein the first, second, third, and fourth translation initiation sequences in each host cell of the plurality of host cells all comprise the sequence (5' to 3') NNNNNATGNGA, where N is C, G, A, or T / U (SEQ ID NO:1).

78. 78. The method of claim 76 or 77, wherein the first, second, third, and fourth polynucleotides are integrated into one or more chromosomes of each eukaryotic host cell of the plurality of eukaryotic host cells.

79. 79. The method of claim 78, wherein the first, second, third, and fourth polynucleotides are integrated into the same chromosomal locus in each eukaryotic host cell of the plurality of eukaryotic host cells.

80. 78. The method of claim 76 or 77, wherein the first, second, third, and fourth polynucleotides are part of one or more extrachromosomal polynucleotides in each eukaryotic host cell of the plurality of eukaryotic host cells.

81. 81. The method of any one of claims 76 to 80, wherein the eukaryotic host cell is a mammalian host cell.

82. 82. The method of claim 81, wherein the mammalian host cell is a Chinese hamster ovary (CHO) cell.

83. 1. A kit of polynucleotides for the expression of a bispecific antibody comprising a first half antibody comprising a first antibody heavy chain and a first antibody light chain, and a second half antibody comprising a second antibody heavy chain and a second antibody light chain, comprising: (a) a first polynucleotide comprising a first translation initiation sequence operably linked to a first open reading frame encoding a first antibody heavy chain; (b) a second polynucleotide comprising a second translation initiation sequence operably linked to a second open reading frame encoding the first antibody light chain; (c) a third polynucleotide comprising a third translation initiation sequence operably linked to a third open reading frame encoding a second antibody heavy chain; (d) a fourth polynucleotide comprising a fourth translation initiation sequence operably linked to a fourth open reading frame encoding a second antibody light chain; Including, one or more of the first, second, third, and fourth translation initiation sequences is not operably linked to its respective open reading frame when present in the naturally occurring host cell genome; when each half antibody is expressed individually in a eukaryotic host cell, the first half antibody is expressed at a lower level than the second half antibody; one or both of the first translation initiation sequence and the second translation initiation sequence are weaker than one or both of the third translation initiation sequence and the fourth translation initiation sequence; kit.

84. 84. The kit of claim 83, wherein the first, second, third, and fourth polynucleotides are part of one or more expression vectors.

85. 85. The kit of claim 83 or 84, wherein the first, second, third, and fourth translation initiation sequences all comprise the sequence (5' to 3') NNNNNATGNGA, where N is C, G, A, or T / U (SEQ ID NO:1).

86. 86. The kit of any one of claims 83 to 85, wherein the first translation initiation sequence comprises the sequence of SEQ ID NO:9, the second translation initiation sequence comprises the sequence of SEQ ID NO:9, the third translation initiation sequence comprises the sequence of SEQ ID NO:2, and the fourth translation initiation sequence comprises the sequence of SEQ ID NO:

11.

87. 87. The kit of any one of claims 83 to 86, wherein each of the first, second, third, and fourth polynucleotides is operably linked to a promoter.

88. 88. The kit of claim 87, wherein the first polynucleotide and the second polynucleotide are operably linked to the same promoter, and the third polynucleotide and the fourth polynucleotide are operably linked to the same promoter.

Images