Methods for selecting clonal cells expressing correctly assembled polyspecific binding molecules

Abstract

The present disclosure relates to methods for screening producer cells for clones expressing a high percentage of correctly assembled polyspecific binding molecules.

Description

[Technical field]

[0001] The present invention relates to methods for selecting cells that express multispecific binding molecules, such as antibodies, particularly bispecific antibodies, that have high levels of correctly assembled proteins.

[0002] Bispecific antibodies have shown a lot of promise as a therapeutic approach. Bispecific antibodies are entering clinical studies in record numbers, mostly developed for cancer. Such molecules, unlike naturally occurring IgG molecules, contain at least two, and usually at least four, different chains (e.g., two heavy chains and two light chains). Thus, there are several different permutations for the assembly of the complete product, whereas for naturally occurring IgG molecules, there is only one.

[0003] To solve these chain pairing problems, several solutions have been proposed. Knobs into holes are commonly used to direct heavy chain-heavy chain pairing. Various techniques, such as the CrossMabs approach, have been used to direct heavy chain-light chain pairing. Nevertheless, there is a need to select clonal producing cell lines that efficiently express bispecifics, where high levels of correct assembly are achieved and correctly paired molecules can be easily separated from incorrectly paired molecules during downstream processing. The present invention fulfills this need.

[0004] Multispecific therapeutic molecules bind to at least two different antigens on cancer cells and recruit at least one effector cell (NK cell or T cell) to the tumor site. Therefore, the protein engineering tools used to generate such composite molecules involve the complex design of multiple polypeptide chains that need to come together in the correct ratio to form a fully functional multispecific molecule. This poses a significant challenge to the cellular machinery.

[0005] This disclosure also describes methods for determining correct assembly of trispecific molecules (e.g., an IgG molecule that binds to HER1 (antigen 1), HER2 (antigen 2) on cancer cells and recruits T cells via an scFv that binds to CD3 (antigen 3) on the T cell receptor), for example, using In-Beacon assays and plate-based assays. Summary of the Invention

[0006] The present invention provides a method that allows individual cell clones to be analyzed in vitro while growing and secreting recombinant protein. By using reagents that specifically bind to the correctly paired regions of the product, different producing clones can be assessed and compared, allowing the selection of cells that efficiently produce an improved ratio of correctly to incorrectly formed products.

[0007] Thus, in a first aspect, the present invention provides a method for selecting cells for expression of a polyspecific binding molecule, comprising the steps of: (a) providing a population of host cells, (i) comprises one or more nucleic acid sequences encoding at least two polypeptides; The polypeptides are expressed by a host cell, wherein a first polypeptide forms a first binding site for a target molecule and a second polypeptide forms a second binding site for the target molecule which is distinct from the first binding site; or (ii) comprises one or more nucleic acid sequences encoding at least four polypeptides; The polypeptide is expressed by a host cell; providing a population of host cells, wherein the first polypeptide and the third polypeptide together form a first binding site for a target molecule, and the second polypeptide and the fourth polypeptide together form a second binding site for the target molecule that is different from the first binding site; (b) contacting the host cell with (i) a first labeled reagent that selectively binds to the first binding site and (ii) a second labeled reagent that selectively binds to the second binding site; (c) measuring the levels of the first and second labeled reagents bound to their respective binding sites; (d) selecting one or more host cells expressing the polyspecific binding molecule based on a comparison of the two levels measured in step (c).

[0008] In another aspect, the invention provides a method for selecting cells for expression of a polyspecific binding molecule, comprising the steps of: (a) providing a population of host cells comprising one or more nucleic acid sequences encoding at least two polypeptides, The polypeptide is expressed by a host cell; providing a population of host cells, wherein a first polypeptide forms a first binding site for a target molecule and a second polypeptide forms a second binding site for the target molecule that is different from the first binding site; (b) contacting the host cell with (i) a first labeled reagent that selectively binds to the first binding site and (ii) a second labeled reagent that selectively binds to the second binding site; (c) measuring the levels of the first and second labeled reagents bound to their respective binding sites; (d) selecting one or more host cells expressing the polyspecific binding molecule based on a comparison of the two levels measured in step (c).

[0009] In a further aspect, the invention provides a method for selecting cells for expression of a polyspecific binding molecule, comprising the steps of: (a) providing a population of host cells comprising one or more nucleic acid sequences encoding at least four polypeptides, The polypeptide is expressed by a host cell; providing a population of host cells, wherein the first polypeptide and the third polypeptide together form a first binding site for a target molecule, and the second polypeptide and the fourth polypeptide together form a second binding site for the target molecule that is different from the first binding site; (b) contacting the host cell with (i) a first labeled reagent that selectively binds to the first binding site and (ii) a second labeled reagent that selectively binds to the second binding site; (c) measuring the levels of the first and second labeled reagents bound to their respective binding sites; (d) selecting one or more host cells expressing the polyspecific binding molecule based on a comparison of the two levels measured in step (c).

[0010] In one embodiment, the third and fourth polypeptides are the same.

[0011] In one embodiment, the host cell comprises one or more identical nucleic acid sequences encoding at least two or at least four polypeptides.

[0012] In one embodiment, the polyspecific binding molecule is secreted by the host cell.

[0013] In one embodiment, the polyspecific binding molecule is a multispecific antibody or a bispecific antibody.

[0014] In one embodiment, the first and / or second labeling reagent comprises a target antigen of the polyspecific binding molecule.

[0015] In one embodiment, the labels of the first and second reagents are different.

[0016] In one embodiment, the first and / or second labeling reagent is a fluorescent labeling reagent.

[0017] In one embodiment, the first and / or second labeling reagent is an anti-idiotypic antibody-fluorophore conjugate.

[0018] In one embodiment, the host cell is a mammalian cell.

[0019] In one embodiment, the host cells are cultured in a volume of about 0.3 nanoliters to about 500 microliters.

[0020] In one embodiment, the host cells are cultured in a microplate or in a fluidic device.

[0021] In one embodiment, the host cells are cultured in a microfluidic device, optionally comprising a microfluidic channel to which multiple isolation pens are fluidly connected, and optionally the host cells are loaded into the microfluidic device such that each of the multiple isolation pens is loaded with one host cell.

[0022] In one embodiment, the method further comprises incubating the host cells selected according to step (d) under conditions that allow for expression and assembly of the at least two different polypeptides into a multispecific binding molecule, and isolating the multispecific binding molecule.

[0023] In one aspect, the invention relates to a method of expressing a polyspecific binding molecule in a host cell, comprising incubating a host cell selected according to the methods described herein under conditions that allow for the expression and assembly of at least two different polypeptides into a polyspecific binding molecule, and isolating the polyspecific binding molecule.

[0024] In a further aspect, the invention relates to polyspecific binding molecules prepared by the methods of expressing the polyspecific binding molecules described herein in a host cell. [Brief description of the drawings]

[0025] [Figure 1] (A) A representation of the positioning of individual isolation pens and optoelectronics in a microfluidic device such as the Berkeley Lights Beacon, and (B) a schematic showing a typical workflow using the Beacon to select cells. The antibodies shown are, by way of example, labeled anti-idiotypic antibodies that bind to correctly formed binding sites within the multispecific antibodies of interest produced by the cell clones in each pen. [Diagram 2] Normalized intensity of varying proportions of "spiked" mAbs in null CCS (cell culture supernatant). Different ratios of monoclonal antibodies mAb1 and mAb2 were spiked into cell culture supernatant. mAb1 and mAb2 were immobilized on protein A-coated wells of a 96-well plate, washed, and the proportion of binding sites was determined by incubation with fluorescently labeled antigen that specifically binds to the binding sites of the monoclonal antibodies. [Diagram 3] Normalized intensity of microplate plate assay to determine the heterogeneity of bispecific antibodies with different architectures. Cell culture supernatants from host cells producing bispecific antibodies of two different architectures (Bs1 and Bs2) were added to protein A-coated wells of a 96-well plate, thereby immobilizing the bispecific antibodies. The antibodies were washed and the ratio of binding sites was determined by incubation with fluorescently labeled antigen that specifically binds to the individual binding sites of the bispecific antibodies. [Figure 4] Schematic workflow of a fluorescence-based microplate plate assay to determine bispecific antibody heterogeneity (Ag1=mAb1-bound antigen 1, Ag2=mAb2-bound antigen 2, FITC=fluorescein isothiocyanate, Bs=bispecific antibody, CCS=cell culture supernatant). For each well, the fluorescence intensity is determined and normalized to each of the reagents (i.e., labeled antigens) to obtain an intensity score for each reagent and well. A pairing score can be calculated by dividing the lowest intensity score by the highest intensity score of each well. Wells containing only fully assembled bispecific antibodies will result in identical intensity scores for each reagent, and therefore a pairing score of 1 (i.e., 100%). Mispairing will result in different intensity scores for each reagent, and therefore a pairing score less than 1, e.g., 0.95, 0.9, or 0.8. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0026] Methods for selecting cells for expression of polyspecific binding molecules - Patents.com In a first aspect, the present invention provides a method for selecting cells for expression of a polyspecific binding molecule, comprising the steps of: (a) providing a population of host cells, (i) comprises one or more nucleic acid sequences encoding at least two polypeptides; The polypeptide is expressed by a host cell; the first polypeptide forms a first binding site for the target molecule and the second polypeptide forms a second binding site for the target molecule which is different from the first binding site; or (ii) comprises one or more nucleic acid sequences encoding at least four polypeptides; The polypeptide is expressed by a host cell; providing a population of host cells, wherein the first polypeptide and the third polypeptide together form a first binding site for a target molecule, and the second polypeptide and the fourth polypeptide together form a second binding site for the target molecule that is different from the first binding site; (b) contacting the host cell with (i) a first labeled reagent that selectively binds to the first binding site and (ii) a second labeled reagent that selectively binds to the second binding site; (c) measuring the levels of the first and second labeled reagents bound to their respective binding sites; (d) selecting one or more host cells expressing the polyspecific binding molecule based on a comparison of the two levels measured in step (c).

[0027] In another aspect, the invention provides a method for selecting cells for expression of a polyspecific binding molecule, comprising the steps of: (a) providing a population of host cells comprising one or more nucleic acid sequences encoding at least two polypeptides, The polypeptide is expressed by a host cell; providing a population of host cells, wherein a first polypeptide forms a first binding site for a target molecule and a second polypeptide forms a second binding site for the target molecule that is different from the first binding site; (b) contacting the host cell with (i) a first labeled reagent that selectively binds to the first binding site and (ii) a second labeled reagent that selectively binds to the second binding site; (c) measuring the levels of the first and second labeled reagents bound to their respective binding sites; (d) selecting one or more host cells expressing the polyspecific binding molecule based on a comparison of the two levels measured in step (c).

[0028] In another aspect, the invention provides a method for selecting cells for expression of a polyspecific binding molecule, comprising the steps of: (a) providing a population of host cells comprising one or more nucleic acid sequences encoding at least four polypeptides, The polypeptide is expressed by a host cell; providing a population of host cells, wherein the first polypeptide and the third polypeptide together form a first binding site for a target molecule, and the second polypeptide and the fourth polypeptide together form a second binding site for the target molecule that is different from the first binding site; (b) contacting the host cell with (i) a first labeled reagent that selectively binds to the first binding site and (ii) a second labeled reagent that selectively binds to the second binding site; (c) measuring the levels of the first and second labeled reagents bound to their respective binding sites; (d) selecting one or more host cells expressing the polyspecific binding molecule based on a comparison of the two levels measured in step (c).

[0029] The first polypeptide, the second polypeptide, the third polypeptide, and the fourth polypeptide may be referred to herein as polypeptide (i), polypeptide (ii), polypeptide (iii), and polypeptide (iv), respectively.

[0030] In some embodiments, the methods described herein are for selecting host cells for expression of correctly assembled polyspecific binding molecules.

[0031] In some embodiments, when properly assembled, the polyspecific binding molecule comprises one copy of each of the first and second polypeptides, or one copy of each of the first, second, third, and fourth polypeptides.

[0032] In some embodiments, at least two polypeptides or at least four polypeptides are heterologous polypeptides. As used herein, a heterologous polypeptide is a polypeptide that is not naturally expressed by the host cell, i.e., a polypeptide that is derived from a different organism or cell type compared to the host cell.

[0033] In some embodiments, the third and fourth polypeptides are the same.

[0034] In some embodiments, the target molecule bound by the first binding site (formed by the first polypeptide, or by the first polypeptide and a third polypeptide) is different from the target molecule bound by the second binding site (formed by the second polypeptide, or by the second polypeptide and a fourth polypeptide).

[0035] In some embodiments, the first binding site is an antigen-binding region of a first immunoglobulin and the second binding site is an antigen-binding region of a second immunoglobulin.

[0036] In some embodiments, at least two polypeptides each comprise three immunoglobulin complementarity determining regions (CDRs). For example, a first polypeptide may comprise a first set of three CDRs, a second polypeptide may comprise a second set of three CDRs, a third polypeptide may comprise a third set of three CDRs, and a fourth polypeptide may comprise a fourth set of three CDRs.

[0037] In some embodiments, the first polypeptide comprises a first immunoglobulin heavy chain Fab region and the second polypeptide comprises a second immunoglobulin heavy chain Fab region. In further embodiments, the third polypeptide comprises a first immunoglobulin light chain Fab region and the fourth polypeptide comprises a second immunoglobulin light chain Fab region. In still further embodiments, the first polypeptide comprises a first immunoglobulin heavy chain Fab region, the second polypeptide comprises a second immunoglobulin heavy chain Fab region, the third polypeptide comprises a first immunoglobulin light chain Fab region and the fourth polypeptide comprises a second immunoglobulin light chain Fab region.

[0038] In some embodiments, at least one of the polypeptides comprises a signal peptide that provides for secretion of the polyspecific binding molecule from a host cell.

[0039] In some embodiments, the polyspecific binding molecule is secreted by the host cell.

[0040] In some embodiments, the polyspecific binding molecule is a multispecific antibody or a bispecific antibody.

[0041] The first and second labeling reagents generally comprise a moiety that selectively binds to a first or second binding site, respectively, and a label. Selective binding of the first and second labeling reagents to the first or second binding site, respectively, generally requires correct assembly of the first or second binding site following expression of the polypeptide by a host cell.

[0042] In some embodiments, the first and / or second labeling reagent comprises a target antigen of the polyspecific binding molecule, i.e., the labeling reagent comprises a target molecule, or a fragment thereof, to which the first and second binding moieties bind. Preferably, the first and second labeling reagents comprise a target antigen of the polyspecific binding molecule.

[0043] In some embodiments, the first and / or second labeling reagent is an anti-idiotypic antibody or antibody fragment (e.g., a Fab or scFv molecule). Preferably, the first and second labeling reagents are anti-idiotypic antibodies or antibody fragments.

[0044] In some embodiments, the labels of the first and second reagents are different.

[0045] In some embodiments, the first and / or second labeling reagent is a fluorescent labeling reagent. Preferably, the first and second labeling reagents are fluorescent labeling reagents.

[0046] In some embodiments, the first and / or second labeling reagent is an anti-idiotypic antibody-fluorophore conjugate. Preferably, the first and second labeling reagents are anti-idiotypic antibody-fluorophore conjugates.

[0047] In some embodiments, the host cell is a mammalian cell.

[0048] In some embodiments, the host cells are cultured in a volume of about 0.3 nanoliters (nL) to about 500 microliters (μL). Preferably, the cells are cultured in a volume of about 0.4 nanoliters to about 250 microliters, more preferably about 0.5 nanoliters to about 200 microliters.

[0049] In some embodiments, the host cells are cultured in microplates.

[0050] In some embodiments, the host cells are cultured in a fluidic device (eg, a microfluidic device such as the Beacon® Optofluidic System from Berkeley Lights).

[0051] In some embodiments, the microfluidic device includes a microfluidic channel to which multiple isolation pens are fluidly connected. Host cells can be loaded into the microfluidic device such that each of the multiple isolation pens is loaded with one host cell.

[0052] multispecific binding molecules The method of the present invention can be used to screen a population of cells expressing polyspecific binding molecules to identify high-performance clones.A polyspecific binding molecule in the context of the present invention is a complex of two or more different polypeptide components, which comprises at least two different binding sites that bind to a target molecule.In other words, the present invention is applicable to molecular complexes, in which there are multiple components that can assemble with each other in different ways, and it is desirable to maximize the correct assembly of the molecule.

[0053] As used herein, a clone is a host cell or population of host cells that is genetically homogeneous.

[0054] Each polypeptide component comprises a binding region of a target molecule of interest. The binding region of each component can pair with the binding region of another component to form a binding site of the target molecule. The overall molecular complex has at least two different binding sites, which can be for different sites on the same target molecule, or more generally, for two different target molecules. Examples of target molecules include cell surface molecules such as receptors, spike proteins, extracellular proteins, or any antigenic protein.

[0055] In one embodiment, the first target molecule (antigen 1) of the polyspecific binding molecule is HER1 and the second target molecule (antigen 2) of the polyspecific binding molecule is HER2 on cancer cells. In a further embodiment, the polyspecific binding molecule recruits T cells via an scFv that binds to CD3 (antigen 3) on the T cell receptor.

[0056] Thus, in one embodiment, the method of selecting cells for expression of a polyspecific binding molecule further comprises incubating host cells selected as described herein under conditions that allow for expression and assembly of at least two different polypeptides into a polyspecific binding molecule, and isolating the polyspecific binding molecule.

[0057] In another aspect, the invention provides a method of expressing a polyspecific binding molecule in a host cell, the method comprising incubating a host cell selected according to the methods described herein under conditions that allow for the expression and assembly of at least two different polypeptides into a polyspecific binding molecule, and isolating the polyspecific binding molecule.

[0058] In another aspect, the invention provides polyspecific binding molecules prepared by the methods of selecting cells for expression of a polyspecific binding molecule or the methods of expressing a polyspecific binding molecule in a host cell described herein.

[0059] In some embodiments, the polypeptide is a single domain antibody.

[0060] In some embodiments, the polyspecific binding molecule is a multispecific antibody or a bispecific antibody, hi some embodiments, the polyspecific binding molecule is an IgG-like bispecific antibody, such as DVD-IgG, IgG-scFv-scFv, scFv4, IgG-Fab, IgG-VH / VL, or DVI-IgG.

[0061] In some embodiments, the target molecule bound by the first binding site (formed by the first polypeptide, or by the first polypeptide and a third polypeptide) is different from the target molecule bound by the second binding site (formed by the second polypeptide, or by the second polypeptide and a fourth polypeptide).

[0062] A typical example of a polyspecific binding molecule is a bispecific antibody, which in contrast to naturally occurring IgG antibodies generally comprises two different heavy chains and two different light chains, so as to have two different antigen-binding regions. In some implementations, a common light chain is used, so that only three different chains are present, i.e., the third and fourth polypeptides may comprise an immunoglobulin light chain Fab region and may be identical.

[0063] Some formats of bispecific antibodies do not contain full-length heavy or light chains, and therefore the polypeptide chains may contain some related immunoglobulin constant region sequences and optionally immunoglobulin antigen-binding regions (i.e., complementarity determining regions (CDRs)). For example, the polypeptide chains may contain the Fab region of an immunoglobulin without any Fc region, which may be omitted or replaced by alternative sequences that provide pairing, such as other types of polypeptides that dimerize (e.g., leucine zippers). Naturally occurring immunoglobulins generally have three CDRs on each polypeptide chain, such that the six CDRs form the antigen-binding site. Thus, each immunoglobulin antigen-binding region of a polypeptide typically contains three CDRs.

[0064] In another embodiment, the polypeptide chain is a complete or substantially complete immunoglobulin chain, e.g., an immunoglobulin light or heavy chain. The sequence can be engineered to enforce correct heavy chain-to-heavy chain pairing by substitutions in the Fc region, e.g., mutations to generate cysteine ​​residues to provide disulfide bonds, "knobs-in-holes" type mutations such as the WSAV approach, and / or substitutions that direct electrostatic interactions (electrostatic steering).

[0065] The sequences can be engineered to enforce correct heavy to light chain pairing by substitutions within the Fab region (constant or variable domains), e.g., mutations generating cysteine ​​residues to provide disulfide bonds, electrostatic steering, and / or domain swapping such as the CrossMAbs approach.

[0066] Other formats exist which provide more than two different binding sites, for example, three or four binding sites, and such approaches are well known in the art.

[0067] Various formats also exist in which additional sequences are added to the immunoglobulin sequence, e.g., IgG-like bispecific antibodies, e.g., DVD-IgG, IgG-scFv-scFv, scFv4, IgG-Fab, IgG-VH / VL, DVI-IgG; as well as fusions, e.g., dock-and-lock (see Bratt et al., 2017, BioProcess International 15(11):36-42).

[0068] host cell Any suitable host cell type may be used in the methods of the invention that can be genetically engineered to express and secrete the polyspecific binding molecules. Preferred host cells are those that can be used to express the polyspecific binding molecules on a large scale for commercial production of the multispecific molecules.

[0069] Preferably, the host cell is not a bacterial cell.

[0070] In some embodiments, the host cell is a eukaryotic cell, such as a mammalian, yeast, or insect cell.

[0071] In one embodiment, the host cell is a mammalian cell. Exemplary species from which the host cell may be derived include human, mouse, rat, Chinese hamster, Syrian hamster, monkey, ape, dog, horse, ferret, and cat.

[0072] In certain embodiments, the mammalian host cell is a Chinese Hamster Ovary (CHO) cell. In one embodiment, the host cell is a CHO-K1 cell, a CHOK1SV cell, a DG44 CHO cell, a DUXB11 CHO cell, a CHO-S, a CHO GS knockout cell, a CHOK1SV FUT8 knockout cell, a CHOZN, or a CHO-derived cell. A CHO GS knockout cell (e.g., a GSKO cell) is, for example, a CHO-K1SV GS knockout cell (Lonza Biologics, plc). A CHO FUT8 knockout cell is, for example, a Potelligent® CHOK1SV FUT8 knockout (Lonza Biologics, plc).

[0073] Other mammalian host cells include HeLa, MDCK, HEK293, HEK293T, HT1080, H9, HepG2, MCF7, Jurkat, NIH3T3, PC12, PER.C6, BHK (baby hamster kidney), VERO, SP2 / 0, NS0, YB2 / 0, Y0, EB66, C127, and COS (e.g., COS1 and COS7).

[0074] In other embodiments, the host cell is a non-mammalian cell, such as an avian, fish, insect, plant, fungal, or yeast cell.

[0075] In embodiments, the eukaryotic cell is a lower eukaryotic cell, such as, for example, a yeast cell (e.g., a species of the genus Pichia (e.g., Pichia pastoris, Pichia methanolica, Pichia kluyveri, and Pichia angusta), the genus Komagataella, the genus Saccharomyces (e.g., Saccharomyces cerevisae, Saccharomyces kluyveri, Saccharomyces uvarum), or the genus Kluyveromyces (e.g., Kluyveromyces lactis, Kluyveromyces marxianus). In some embodiments, the eukaryotic cell is a Pichia pastoris species. Examples of Pichia pastoris strains include, but are not limited to, X33, GS115, KM71, KM71H, and CBS7435.

[0076] In embodiments, the eukaryotic cell is an insect cell (e.g., an Sf9, Mimic™ Sf9, Sf21, High Five™ (BT1-TN-5B1-4), or BT1-Ea88 cell).

[0077] Suitable host cells are commercially available, for example from culture collections such as DSMZ (Deutsche Sammlung von Mikroorganismen and Zellkulturen GmbH, Braunschweig, Germany) or the American Type Culture Collection (ATCC).

[0078] A population of host cells transformed with nucleic acid sequences that express polyspecific binding molecules. Nucleic acid sequences encoding the different components of the polyspecific binding molecule can be introduced into a population of host cells using techniques well known in the art. The sequences encoding the constituent polypeptide chains, operably linked to regulatory control elements that drive expression of the polypeptides in the host cells, are typically present in one or more nucleic acid vectors. One or more of the polypeptides also typically includes a signal sequence that directs secretion of the polypeptide from the host. The vector typically includes one or more selectable markers to allow for the selection of host cells that have incorporated the nucleic acid vector. Examples of selectable markers include amino acid auxotrophy-based markers such as dhfr and glutamine synthetase (GS).

[0079] After cell selection for the presence of the introduced nucleic acid sequences into the host cells, a cell population can be generated in which a plurality of cells comprises one or more nucleic acid sequences encoding (i) a first polypeptide comprising a first immunoglobulin antigen binding region, such as an immunoglobulin CDR, (ii) a second polypeptide comprising a second immunoglobulin antigen binding region, such as an immunoglobulin CDR, (iii) a third polypeptide comprising a third immunoglobulin antigen binding region, such as an immunoglobulin CDR, and (iv) a fourth polypeptide comprising a fourth immunoglobulin antigen binding region, such as an immunoglobulin CDR, wherein the first and third immunoglobulin antigen binding regions (e.g., CDRs) together form a first antigen binding site and the second and fourth immunoglobulin antigen binding regions (e.g., CDRs) together form a second antigen binding site that is distinct from the first antigen binding site.

[0080] The goal of the selection process is to identify and develop further specific clones that produce high levels of the correctly paired molecular complex (e.g., polyspecific binding molecule) of interest. Various genetic factors may mean that in a population of clones transformed with the same sequence, not all clones will function in the same manner.

[0081] The selection process is not based on testing the binding of different sequences where different cells have different variants of the polypeptide components, but rather, the cells within the population / pool have all been transformed with the same set of sequences.

[0082] Thus, the host cells in the population of host cells used in the method according to the invention contain the same one or more nucleic acid sequences encoding the polypeptides described herein. In one embodiment, the host cells used in the method according to the invention contain the same one or more nucleic acid sequences encoding at least two polypeptides. In another embodiment, the host cells used in the method according to the invention contain the same one or more nucleic acid sequences encoding at least four polypeptides.

[0083] For example, the host cells are all transformed with the same one or more nucleic acid sequences. In one embodiment, providing a population of host cells comprising one or more nucleic acid sequences comprises introducing the one or more nucleic acid sequences into the host cells. In certain embodiments, providing a population of host cells comprising one or more nucleic acid sequences further comprises selecting host cells for successful introduction of the one or more nucleic acid sequences into the host cells.

[0084] The mutation originates from the host cell itself. Thus, the population of cells to be tested and selected contains the same sequences encoding the first, second, and, if applicable, the third and fourth polypeptides. Due to transformation efficiency, it is possible that not all cells in the population contain all the sequences, but in any case such cells will not be of interest.

[0085] The CDRs may be a portion of an immunoglobulin Fab region, and thus the plurality of cells may comprise (i) a first polypeptide comprising a first immunoglobulin heavy chain Fab region, (ii) a second polypeptide comprising a second immunoglobulin heavy chain Fab region, (iii) a third polypeptide comprising the first immunoglobulin light chain Fab region, and (iv) a fourth polypeptide comprising the second immunoglobulin light chain Fab region. The first immunoglobulin heavy chain Fab region and the first immunoglobulin light chain Fab region together form a first immunoglobulin antigen-binding region, and the second immunoglobulin heavy chain Fab region and the second immunoglobulin light chain Fab region together form a second immunoglobulin antigen-binding region that is distinct from the first immunoglobulin antigen-binding region.

[0086] Further, as noted above, some approaches involve a common chain such that there are three different polypeptides rather than four or more. Thus, one of the polypeptides (i)-(iv) can be identical to one or the other. For example, the third and fourth polypeptides can be identical, and there are in fact three different polypeptides, a first, a second and a third polypeptide, such that the first polypeptide pairs with the third polypeptide at the first antigen binding site and the second polypeptide also pairs with the third polypeptide at the second antigen binding site.

[0087] The polypeptide may also contain an Fc region to form a full-length heavy and light chain.

[0088] As discussed above, different regions of the polypeptide may be modified to promote heavy chain heavy chain pairing and / or heavy chain light chain pairing.

[0089] Screening a population of host cells in a suitable device Host cells to be screened for expression of correctly assembled polyspecific binding molecules by the methods of the invention can be cultured in a variety of suitable formats or devices known to those of skill in the art, including microplates or fluidic devices (e.g., microfluidic or nanofluidic devices).

[0090] Selection of one or more cells expressing the polyspecific binding molecules based on a comparison of the levels of labeling reagent bound to the respective binding sites is generally facilitated by loading individual cells into compartments of a device used for culturing cells (e.g., individual wells of a microplate or an isolation pen of a fluidic device) such that, after clonal expansion, each compartment contains a population of cells resulting from the clonal expansion of a single cell. Thus, in one embodiment, host cells are loaded into compartments of a device used for culturing cells (e.g., a microplate or a fluidic device) such that a plurality of compartments are each loaded with one host cell. The compartments can be wells of a microplate or an isolation pen of a fluidic device.

[0091] Preferably, the host cells are clonally propagated to obtain a plurality of host cell populations, each of which is genetically homogeneous. Thus, in one embodiment of the method disclosed herein, providing a population of host cells comprises clonally propagating the host cells to obtain a plurality of host cell populations, each of which is genetically homogeneous. In a preferred embodiment, the host cells are loaded into compartments of a device (e.g., a microplate or a fluidic device) used for culturing cells, such that each of the compartments is loaded with one host cell, and the host cells are then clonally propagated to obtain a plurality of host cell populations, each of which is genetically homogeneous.

[0092] In one embodiment, the host cells are cultured in a microplate (also called a multiwell plate or a microtiter plate), such as a 24-well plate, a 48-well plate, a 96-well plate, a 384-well plate, or a 1536-well plate. In a preferred embodiment, the host cells are cultured in a 96-well plate or a 384-well plate, more preferably a 96-well plate.

[0093] In further embodiments, the host cells are cultured in a fluidic device (eg, a microfluidic or nanofluidic device).

[0094] In certain embodiments, a population of cells is introduced into a fluidic device (such as a microfluidic or nanofluidic device) that includes a channel (such as a microchannel or nanochannel) to which multiple isolation pens are fluidically connected. The fluidic device includes a substrate, and the channel and pens are part of a fluidic structure disposed on a surface of the substrate. Cells suspended in a liquid medium can flow along the channel and pass through the isolation pens. The device is configured to allow individual cells to be loaded into the isolation pens such that each isolation pen contains only one cell, and also to allow cells in a particular isolation pen to be released into the channel for collection. The pens can be from about 0.3 nanoliters to about 500 microliters in volume, e.g., in a microfluidic device, from about 0.3 to about 10 nL in volume.

[0095] The isolation pen can include a fluid isolation structure having a separation region with a single opening and a connection region fluidly connecting the separation region to the channel, the connection region including a proximal opening to the channel.

[0096] The substrate may be tilted at a small angle from the horizontal so that the cells sink to the bottom of the isolation pen and away from the narrow single opening to the channel.

[0097] One example of a microfluidic device technology to achieve this is the use of OptoElectric Positioning (OEP), such as the Berkeley Lights Inc., Emeryville, CA Beacon system, as described, for example, in US 9,857,333, Mocciaro et al., 2018, Light-activated cell identification and sorting (lacis) for selection of edited clones on a nanofluidic device. Commun Biol. 1:41, and Le et al., 2018, A novel mammalian cell line development platform utilizing nanofluidics and optoelectro positioning technology. Biotechnol Prog. 34:1438-46. OEP is based on a microfluidic device that includes transparent electrodes on a silicon substrate with a fluid chamber sandwiched between the two. The substrate is fabricated with an array of photosensitive transistors. When focused light hits the transistors and a voltage is applied, a non-uniform electric field is generated. It uses light-induced OEP to impart a repulsive negative dielectrophoretic (DEP) force to particles (including cells) (Figure 1a). In the absence of targeting light, no force is generated. When light is irradiated onto the photoconductive material, a DEP force is generated and cells trapped in the optical "cage" can move across the chamber. In addition, isolation pens (referred to as NanoPens™ in Berkeley Lights devices) are integrated into the chip to isolate cells from each other and allow on-chip culture of well-separated colonies emanating from single cells. After characterization, selected clones can be exported from the microfluidic device for further processing. Export is the reverse of the import process, where the desired cells are moved using OEP from a single isolation pen into the main channel and flushed, for example, into target wells of a 96-well plate placed in a CO2- and temperature-controlled incubator.

[0098] The microfluidic device can be pre-wetted with a suitable wetting solution according to any manufacturer's instructions to create an environment compatible with the host cells and allow good Penning efficiency. The device can then be primed with a cell culture medium suitable for host cell growth and protein expression.

[0099] Once the cells are loaded into a suitable device (e.g., a microplate or a fluidic device) and placed into individual compartments (e.g., wells or isolation pens), the cells are incubated to allow for cell growth and clonal expansion and production of the polyspecific binding molecules. Monitoring of the different compartments can be used to ensure monoclonality in each compartment and also to ensure that cells do not overfill the compartments prior to analysis of the polyspecific binding molecules.

[0100] The cells are then analyzed by introduction into the device (e.g., a microplate or fluidic device) of a reagent that specifically binds to the correctly formed first binding site or the correctly formed second binding site in the multispecific molecule. In one embodiment, the reagent comprises a binding site, e.g., a target antigen corresponding to the target molecule or a fragment thereof. In another embodiment, the reagent is an anti-idiotypic antibody, including a fragment thereof, such as a Fab, scFv molecule, specific for one of the correctly formed binding sites in the multispecific molecule. Other reagents include aptamer-oligonucleotide or peptide molecules with the required binding specificity.

[0101] Each binding molecule must be specific for a different binding site so that binding can be differentiated.

[0102] The reagent is typically labeled with a detectable label, allowing the binding of the reagent to its target to be measured in situ within the device. In one embodiment, the label is a chemiluminescent label. In another embodiment, the label is a fluorescent label, such as a fluorophore or a fluorescent protein. In one embodiment, the detectable label is fluorescein or a derivative thereof (e.g., fluorescein isothiocyanate). Exemplary fluorescent proteins are blue fluorescent proteins, such as BFP and mTagBFP, cyan fluorescent proteins, such as ECFP and TagCFP, green fluorescent proteins, such as EGFP and ZsGreen, yellow fluorescent proteins, such as EYFP and ZsYellow, red fluorescent proteins, such as mRFP and mCherry, and far-red proteins, such as E2-Crimson.

[0103] Each different reagent may be labeled with the same or a different detectable label. Preferably, each different reagent is labeled with a different detectable label.

[0104] Each reagent can be introduced into the device separately or simultaneously (this may depend on whether different labels are used and the ability of the imaging system to distinguish between different labels, such as different fluorescent (or chemiluminescent) signals, to allow simultaneous measurements).

[0105] In one embodiment, a first reagent is introduced into the device such that it can enter a compartment (e.g., a well or isolation pen) and contact the multispecific molecule produced by the host cells in the compartment. The reagent is present in the compartment for a period of time (e.g., 45-60 minutes) to provide sufficient binding to the multispecific molecule. After a suitable washing step, if desired, binding of the first reagent to the first binding site in the molecule is determined, for example, by fluorescent imaging of the device. A washing step is then used to remove the first reagent, and the process is repeated with the second reagent, and so on.

[0106] The extent of binding of the first and second reagents is determined, for example, using an image analysis algorithm or script, which can be used to determine an intensity score (e.g., normalized signal intensity) for each compartment (e.g., well or isolation pen) of each reagent, adjusted as necessary to the actual amount of binding depending on the performance of the label used, so that an accurate comparison of the intensity scores of the different reagents can be made.

[0107] In one embodiment, the intensity score is determined by normalizing each level (e.g., signal intensity) of the first and second labeling reagents to a standard. In one embodiment, the standard is a titration curve of each labeling reagent over the relevant concentration range. In one embodiment, the standard is a sample representing a correctly assembled polyspecific binding molecule (e.g., a purified correctly assembled polyspecific binding molecule). In another embodiment, a separate standard is used for each labeling reagent, and each standard is a sample representing a correct assembly of binding sites selectively bound by the respective labeling reagent. For example, the levels measured for the first labeling reagent are normalized using a standard that is a sample representing a correct assembly of a first binding site selectively bound by the first labeling reagent, and the levels measured for the second labeling reagent are normalized using a standard that is a sample representing a correct assembly of a second binding site selectively bound by the second labeling reagent.

[0108] The intensity scores of each reagent are then compared to obtain a pairing score, e.g., a percentage obtained by dividing the lowest intensity score by the highest intensity score of each pen. For example, a similar score of 90% or 95% or more indicates a high level of correctly paired strands, since similar amounts of correctly formed first and second binding sites in the multispecific molecule are present in the compartments (e.g., wells or isolated pens). Expressed differently, if the intensities are within + / -20%, such as + / -10% or + / -5% of each other, this is considered a similar score. Conversely, a score outside of the similar score indicates a significant level of unpaired molecules, e.g., greater than + / -20% with reference to the above calculation. Thus, in one embodiment, levels of the first and second labeling reagents bound to their respective binding sites that are within + / - 20% of each other (i.e., within + / - 20% of each other), preferably within + / - 10% of each other, and most preferably within + / - 5% of each other, are indicative of a correctly paired polyspecific binding molecule (i.e., a polyspecific binding molecule that includes both a first binding site for a target molecule and a second binding site for a target molecule).

[0109] In one embodiment, selecting one or more host cells expressing a polyspecific binding molecule based on a comparison of the two levels (e.g., signal intensity) measured for the first labeling reagent and the second labeling reagent bound to their respective binding sites comprises selecting one or more host cells according to the level of correctly assembled polyspecific binding molecules. In other words, selecting one or more host cells expressing a polyspecific binding molecule comprises ranking the host cells according to the level (e.g., signal intensity) ratio of the first labeling reagent and the second labeling reagent. In this process, host cells with a level (e.g., signal intensity) ratio close to 1 are generally preferred. In certain embodiments, one or more host cells are selected according to their pairing score(s), e.g., one or more host cells determined to exhibit the highest pairing score(s) are selected. In one embodiment, the pairing score is obtained by comparing the levels (e.g., intensity scores) of the first labeling reagent and the second labeling reagent.

[0110] Typically, cells in a particular compartment are scored for productivity (total level of multispecific molecule production), since being able to produce high titers of the product of interest is advantageous for industrial production of cells. Scoring can be a relative score between various clones. The Beacon system offers SpotLight™ human Fc and kappa light chain assays. These are fluorophores that bind to the Fc and / or kappa light chain constant regions and obtain a fluorescent signal that is directly proportional to the amount of antibody expressed in that particular clone. A "score" value is created by measuring the change in fluorescence in or near the neck of the pen (in what is called a diffusion gradient assay). The higher the value of the gradient, the higher the titer. The maximum score is normalized to 100 and the minimum score to 0. Values ​​in between are percentages of the total, and the score is calculated from these values. The rQp is the relative production per cell of the last DiGr assay, i.e., the score divided by the number of cells.

[0111] Thus, in one embodiment, selecting one or more host cells expressing a multispecific binding molecule based on a comparison of the two measured levels of the first labeling reagent and the second labeling reagent bound to their respective binding sites further comprises selecting one or more host cells according to their productivity (total level of multispecific molecule production). In a particular embodiment, the productivity is determined by determining a relative score between multiple clones. The relative score can be a signal (e.g., a fluorescent signal) that represents the amount of multispecific molecule expressed in that particular clone. In one embodiment, the signal is a fluorescent signal determined using a fluorophore that binds to the Fc and / or kappa light chain constant region and obtains a fluorescent signal that is directly proportional to the amount of antibody expressed in that particular clone.

[0112] Relative scores can be normalized to a standard, for example, to the clone with the highest absolute productivity. In one embodiment, to generate relative scores, signals (e.g., fluorescent signals) are normalized by dividing the signal determined for each clone by the highest signal determined for any of the clones. Thus, the clone with the highest productivity can be assigned a productivity of 100%, and the clone with lower productivity can be assigned a corresponding lower value (e.g., 90%, 70%, 50%, etc.).

[0113] Individual clones that meet the thresholds for correct pairing, and typically for productivity, can then be exported from the device, for example to 96-well plates, for further growth. Selected clones are then typically further evaluated to identify the highest producing clones (high Qp).

[0114] Clones can optionally be subjected to further analysis to confirm high levels of correct pairing. For example, limited digestion (e.g., with Lys-C) and evaluation of purified molecules by mass spectrometry (LC-MS) can be used to confirm the levels and structure of mispaired variants for more detailed quantification.

[0115] Apply Selected Clone Selected clones can then be used to establish stable cell lines for use in the production of recombinant polyspecific binding molecules of interest, e.g., at a scale of greater than 500 g per batch, e.g., in bioreactors having a volume of at least 10 L, such as at least about 100, 200, 500, 1000, or 2000 L.

[0116] The invention is further illustrated with reference to the following examples, which are non-limiting. EXAMPLES

[0117] Example 1: Determining mAb heterogeneity using the microfluidic Beacon system material A stable pool of cells expressing the Antibody1 / Antibody2 bispecific antibody Anti-Antibody 1 idiotype antibody conjugated with fluorophore A Antigen 1 conjugated with fluorophore A Anti-Antibody 2 idiotypic antibody conjugated with fluorophore B Antigen 2 conjugated with fluorophore B

[0118] method 1. Prepare the chip (wet the Beacon™ OptoSelect™ chip with the wetting solution supplied by the manufacturer to ensure good penning efficiency and create a cell-friendly environment that allows cell growth) - Day 1 2. Calibration of the assay: Diffusion gradient (DiGr) assay (protein yield), reference images of Fluorophore A and Fluorophore B are obtained for each OptoSelect chip - Day 1 3. Priming Chips and Lines with Cell Culture Medium - Day 2 4. Loading the cells: Loading the plate from cells into specific Optoselect chips - Day 2 5. Growing Cultures – Morning of Day 2 to Day 5 6. Assay: Day 5 to Day 7 a. Productivity assay (DiGr), followed by the shortest time it takes for the fluorophore to diffuse out of the chip and disappear, then b. Fluorophore A assay (fluorophore conjugated to anti-idiotypic antibody A / fluorophore conjugated to antigen A), followed by the shortest time it takes for the fluorophore to diffuse off the chip and disappear, then c. Fluorophore B assay (fluorophore conjugated to anti-idiotypic antibody B / fluorophore conjugated to antigen B). Fluorescent imaging of the chips is performed by the Beacon system, which is capable of capturing brightfield and fluorescent images over multiple time points. Analysis of the images using a script results in assigning an "intensity score" to each NanoPen™ based on the gradient of intensity in a particular area of ​​the NanoPen. The script then compares the intensity scores of fluorophore A and fluorophore B for each NanoPen and generates a "pairing score" based on the similarity of the levels of bound fluorophores; the higher the score, the more similar the levels of bound fluorophore A and B and the higher the percentage of correctly paired heterodimeric antibodies produced by a particular clone. Software to analyze the fluorescent images also includes free software such as ImageJ (available from the National Institutes of Health website) that measures intensity. 7. Export clones based on heterodimerization % and productivity. This can be based either on scoring information generated and stored in the Beacon system and used as default for clone export (e.g., top scoring clones for export are those with higher heterodimerization % + higher productivity) or on the best clone identification information uploaded to the Beacon system. 8. Growth and progression of clones in 96-well plates.

[0119] This procedure can also be performed with a single fluorophore, as measurements can be made sequentially and the first reagent washed away before the second reagent is introduced.

[0120] Example 2: Determination of heterogeneity by varying the ratio of "spiked" mAbs in null CCS (cell culture supernatant) material Protein A coated 96-well plate DPBS+0.05% P20 (Dulbecco's phosphate-buffered saline containing 0.05% (v / v) polysorbate 20 as a surfactant) mAb1 mAb2 FITC-conjugated mAb1-bound antigen 1 (FITC = fluorescein isothiocyanate) FITC-conjugated mAb2-binding antigen 2

[0121] method 1- Equilibrate Protein A coated 96 well plates with 200 μL per well of DPBS+0.05% P20 for 10 min. 2-Remove the buffer and add different ratios of mAb1 and mAb2 spiked in null CCS. 3- Incubate at room temperature for 30-35 minutes and remove the supernatant. 4- Wash wells carrying samples 3 times with 200 μL of DPBS+0.05% P20 per well. 5-Add 100 μL of DPBS+0.05% P20 to each well. 6-1.4 μL of FITC-conjugated mAb1-binding antigen 1 and FITC-conjugated mAb2-binding antigen 2 are added to each well and incubated for an additional 30 minutes. 7-Wash wells 3 times each with 200 μL of DPBS+0.05% P20. 8-100 μL of DPBS+0.05% P20 is added to each well and the fluorescence intensity is measured using a SpectraMax M5e plate reader (Ex / Em=488 / 535 nm was used to measure fluorescence. Settings selected were 50 flashes per reading. All measurements were performed at 25° C. Data was exported and analyzed in Microsoft Excel).

[0122] Data analysis The mean and standard deviation (SD) of triplicate measurements were obtained. The fluorescence intensity obtained from the null-CCS was subtracted from that obtained from the remaining samples. The intensity obtained with FITC-conjugated mAb1 binding antigen 1 was normalized to mAb1, and the intensity obtained with FITC-conjugated mAb1 binding antigen 2 was normalized to mAb2 (Tables 1 and 2). The normalized intensities were plotted as a bar graph (Figure 2). [Table 1] [Table 2]

[0123] The data show that the plate assay can be used to assess the degree of heterogeneity in various CCS samples.

[0124] Example 3. Fluorescence-based 96-well plate assay to determine heterogeneity of bispecific antibodies with different architectures. material Protein A coated 96-well plate DPBS+0.05% P20 Supernatant from cell cultures expressing the unique bispecific antibody bs1 (bispecific architecture a, consisting of mAb1 and mAb2) Supernatant from cell cultures expressing the unique bispecific antibody bs2 (bispecific architecture b, consisting of mAb1 and mAb2) FITC-conjugated mAb1-binding antigen 1 FITC-conjugated mAb2-binding antigen 2

[0125] method 1- Equilibrate Protein A coated 96 well plates with 200 μL per well of DPBS+0.05% P20 for 10 min. 2- Remove the buffer and add 50 nM of sample in CCS to the appropriate wells. The concentration of the sample was adjusted by diluting the stock solution with null CCS (day 4). 3- Incubate at room temperature for 30-35 minutes and remove the supernatant. 4- Wash wells carrying samples 3 times with 200 μL of DPBS+0.05% P20 per well. 5-Add 100 μL of DPBS+0.05% P20 to each well. 6-1.4 μL of FITC-conjugated mAb1-binding antigen 1 and FITC-conjugated mAb2-binding antigen 2 are added to each well and incubated for an additional 30 minutes. 7-Wash wells 3 times each with 200 μL of DPBS+0.05% P20. 8-100 μL of DPBS+0.05% P20 is added to each well and the fluorescence intensity is measured using a SpectraMax M5e plate reader (Ex / Em=488 / 535 nm was used to measure fluorescence. Settings selected were 50 flashes per reading. All measurements were performed at 25° C. Data was exported and analyzed in MS Excel).

[0126] Data analysis The mean and SD of triplicate measurements were obtained. The fluorescence intensity obtained from the null-CCS was subtracted from that obtained from the remaining samples. The intensity obtained for FITC-conjugated mAb1 binding antigen 1 (Ag1-FITC) was normalized to mAb1, and the intensity obtained for FITC-conjugated mAb1 binding antigen 2 (Ag2-FITC) was normalized to mAb2 (Table 3). The normalized intensities were plotted as a bar graph (Figure 3).

[0127] A plate-based format to distinguish the degree of heterogeneity in bispecific cultures is feasible. When unique bs molecules are tested using this method, the intensity ratios obtained for FITC-conjugated mAb1 binding antigen 1 and FITC-conjugated mAb2 binding antigen 2 correlate with the LCMS data obtained with those particular molecules.

[0128] The workflow of the method is shown diagrammatically in FIG. [Table 3]

[0129] Example 4: In-Beacon Assay Single cells carrying the necessary nucleotide sequences to express the trispecific molecule are loaded into individual pens and grown for 4 days by perfusing culture medium into the chip. On day 4, fluorescently labeled antigens 1, 2, and 3 are added to the chip at a K >10-fold greater than the interaction. D The prediction of trispecific expressing clones can be determined by two workflows. (a) Sequential workflow: Fluorescently labeled antigen 1 is first perfused and the fluorescence intensity (after subtraction of the baseline and normalization by the number of cells in each pen) is recorded. Media is perfused to remove antigen 1 and the clones are allowed to grow. Then fluorescently labeled antigen 2 is perfused and the fluorescence intensity (after subtraction of the baseline and normalization by the number of cells in each pen) is recorded. Media is perfused to remove antigen 2 and the clones are allowed to grow. Finally, fluorescently labeled antigen 3 is perfused and the fluorescence intensity (after subtraction of the baseline and normalization by the number of cells in each pen) is recorded. Media is perfused to remove antigen 3 and the clones are allowed to grow. (b) Multiplexed workflow: Antigens 1, 2, and 3 labeled with three different fluorophores with minimal spectral overlap were multiplexed with a K >10-fold higher than the interaction. D Prepare a working concentration of 100 mM NaCl. Mix the fluorescently labeled antigens in an equimolar ratio and perfuse into the chip. Collect the fluorescent signal from each labeled antigen and use the data to estimate the ratio of antigen-binding sites in molecules secreted by the clones. The multiplexing method can only be used if the antigen-binding sites are not within FRET distance (0.2-0.9 nm).

[0130] Example 5: Plate-based assay Plate-based assays can be used to assess the degree of multispecific binding using both sequential and multiplexed approaches. In the plate-based assay, IgG trispecifics are immobilized on protein A-coated 96-well plates and excess molecules are washed off the plate using an appropriate buffer. In both sequential and multiplexed formats, the workflow is similar to that of Beacon described above, except that a fluorescent plate reader is used to measure fluorescence intensity. *************

[0131] The disclosures of each and every patent, patent application, and publication cited herein are incorporated herein by reference in their entirety. Although the present invention has been disclosed with reference to certain aspects, it is apparent that other aspects and variations of the present invention may be devised by those skilled in the art without departing from the true spirit and scope of the present invention. Features and embodiments in different sections may be combined mutatis mutandis.

Claims

1. 1. A method for selecting cells for expression of a polyspecific binding molecule, comprising: (a) providing a population of host cells, (i) comprises one or more nucleic acid sequences encoding at least two polypeptides; the polypeptide is expressed by the host cell; the first polypeptide forms a first binding site for the target molecule and the second polypeptide forms a second binding site for the target molecule which is distinct from the first binding site; or (ii) comprises one or more nucleic acid sequences encoding at least four polypeptides; the polypeptide is expressed by the host cell; providing a population of host cells, wherein the first polypeptide and the third polypeptide together form a first binding site for a target molecule, and the second polypeptide and the fourth polypeptide together form a second binding site for the target molecule that is distinct from the first binding site; (b) contacting the host cell with (i) a first labeled reagent that selectively binds to the first binding site, and (ii) a second labeled reagent that selectively binds to the second binding site; (c) measuring the levels of the first and second labeled reagents bound to their respective binding sites; (d) selecting one or more host cells expressing a polyspecific binding molecule based on a comparison of the two levels determined in step (c).

2. The method of claim 1 (a) (ii), wherein the third and fourth polypeptides are the same.

3. The method of claim 1 or 2, wherein the host cell comprises one or more identical nucleic acid sequences encoding the at least two or at least four polypeptides.

4. The method of any one of claims 1 to 3, wherein the polyspecific binding molecule is secreted by the host cell.

5. The method of any one of claims 1 to 4, wherein the polyspecific binding molecule is a multispecific antibody or a bispecific antibody.

6. The method of any one of claims 1 to 5, wherein the first and / or second labeling reagent comprises a target antigen of the polyspecific binding molecule.

7. The method of any one of claims 1 to 6, wherein the labels of the first and second reagents are different.

8. The method according to any one of claims 1 to 7, wherein the first and / or second labelling reagent is a fluorescent labelling reagent.

9. The method of any one of claims 1 to 8, wherein the first and / or second labeling reagent is an anti-idiotypic antibody-fluorophore conjugate.

10. The method according to any one of claims 1 to 9, wherein the host cell is a mammalian cell.

11. 11. The method of any one of claims 1 to 10, wherein the host cells are cultured in a volume of about 0.3 nanoliters to about 500 microliters.

12. The method according to any one of claims 1 to 11, wherein the host cells are cultured in a microplate or in a fluidic device.

13. 13. The method of claim 12, wherein the host cells are cultured in a microfluidic device, optionally comprising a microfluidic channel to which a plurality of isolation pens are fluidly connected, and optionally the host cells are loaded into the microfluidic device such that each of the plurality of isolation pens is loaded with one host cell.

14. 14. The method of any one of claims 1 to 13, further comprising incubating the host cells selected according to step (d) under conditions that allow expression and assembly of at least two different polypeptides into a polyspecific binding molecule and isolating said polyspecific binding molecule.

15. 15. A polyspecific binding molecule prepared by the method of claim 14.

Images