Selection for the developability of polypeptide drugs in eukaryotic cell display systems.

Abstract

Using the surface display level of a binder (e.g., antibody, receptor) on in vitro cultured higher eukaryotic cells as a predictor of binder developability properties, such as solubility. Display libraries of higher eukaryotic cells, such as mammalian cells, adapted for use in screening surface-displayed binders for developability and affinity for target binding. High-throughput screening of display libraries with inherent selection for developability, including binder solubility, ability to be formulated at high concentrations, low tendency for nonspecific binding, and half-life. Enriching a population of binders for developability properties and / or other properties, such as target binding and affinity, by controlling cell surface display of the binder from an inducible promoter operably linked to the binder-encoding DNA. [Selection diagram] None

Description

[Technical Field]

[0001] The present invention relates to identifying candidate polypeptide drugs that have desirable developability characteristics such as solubility, ability to be formulated at high concentrations, low tendency for non-specific binding, and optimal half-life. The present invention further relates to screening of binding agents in drug discovery, including antibody discovery. [Background technology]

[0002] Antibodies have proven to be an extremely successful drug class, with over 70 therapeutic antibodies approved to date and many more in the development pipeline. However, the manufacture and formulation of polypeptide drugs, such as antibodies, is in many ways a more complex endeavor than small molecule drugs. Several factors affect the practicality of developing polypeptide drugs, such as antibodies, and influence whether a potential antibody candidate will be successfully developed into a drug that is not only effective but also manufacturable, stable, and safe. These factors include chemical stability (e.g., resistance to fragmentation, deamidation, oxidation, and isomerization), physical stability (e.g., conformational stability, tendency of the protein to unfold, aggregation and / or precipitation, colloidal stability), and solution properties (e.g., solubility, tendency to reversibly self-associate in solution, and viscosity at high concentrations). The ability of a polypeptide to be expressed at high yields in cell culture is also a relevant consideration, as the amount of polypeptide secreted from host cells determines the yield of product that can be recovered from the medium. The immunogenicity of an antibody or polypeptide drug is also a consideration. All these factors are collectively referred to as the "developability" characteristics of a product. They are important considerations because they affect the cost and practicality of manufacturing a product, its safety profile, dosing schedule, and mode of administration. When developability issues arise, they can affect the availability or commercial success of the antibody and can cause potential molecules to fail during the development process. Aspects of antibody drug developability, and how they can be measured, have been reviewed. 1~3 .

[0003] It is desirable to produce stable, soluble proteins at high concentrations for administration to patients. For example, antibody concentrations of 50 mg / ml or greater are required for subcutaneous administration, where a relatively large dose must be administered in a small volume. An ideal polypeptide for therapeutic use is one that is highly soluble in aqueous buffer solutions and can therefore be concentrated to high concentrations. Successful production of soluble proteins at high concentrations depends, at least in part, on the polypeptide's resistance to self-association, which can adversely lead to increased viscosity and / or precipitation. Once the solubility limit of a molecule is reached, further increases in the dissolved concentration are no longer possible, resulting in undesirable effects such as the molecule precipitating from solution. Approaching the solubility limit can result in the solution becoming viscous and / or the dissolved polypeptide can self-associate (reversibly or irreversibly) in solution, complicating solution handling and formulation. Worse yet, aggregate formation by poorly soluble and / or unstable polypeptides can increase the risk of immunogenicity when the product is administered to patients. 4、5 Anti-drug immune responses (e.g., anti-drug antibodies) can neutralize the therapeutic effect, and hypersensitivity can result in illness or death.

[0004] The ability to express a polypeptide at high levels can affect the economics of production. In the early stages of antibody discovery, yields from optimized transient expression in mammalian cell culture can reach 50 μg / ml or more. Transient expression is a temporary expression system in which a plasmid encoding the product of interest is transfected into cells, resulting in expression for a period of time, typically declining after 2–7 days as the plasmid is lost from the cells. Stable expression in recombinant cells for drug manufacturing requires stable integration of the coding DNA into the cell genome. Once a stable manufacturing cell line is created, the cells can be cultured for one or two weeks to obtain high yields, which can be approximately 1 mg / ml or more.

[0005] Low yields from transient expression can indicate potential developability issues for drug candidates. For example, the yield of an aggregation-prone anti-angiopoietin antibody ("Ang2") was reported to be only 10 μg / ml, whereas an optimized variant of the antibody yielded 260 μg / ml. 6 However, even when good yields are obtained from transient expression (e.g., >50 μg / ml) and / or stable cell lines (e.g., >1 mg / ml), biophysical problems can arise when polypeptides are enriched above 1 mg / ml. Dobson et al. (2016) 7 ) found that the anti-NGF antibody MEDI1912 had significant biophysical problems when compared to its parent antibody MEDI576, which were not reflected in the expression levels reported from transient cultures of approximately 200 μg / ml. Thus, yield from expression in cell culture is only one aspect of developability and may not be predictive of other important aspects.

[0006] Over the past few decades, display technologies have enabled the creation of large and diverse populations of antibodies and other proteins and peptides from which individual variants with desired target binding properties can be isolated. Desired binding properties include binding to appropriate epitopes on targets with appropriate specificity (e.g., for orthologs and paralogs) and appropriate affinity (e.g., to inhibit receptor-ligand interactions). Display technologies can aid in discovering binders with desired properties by allowing some such aspects to be enriched during selection. For example, by using limited amounts of antigen to drive selection, antibodies with higher affinity can be selected. Deselection against binding to undesired paralogs has also been described. In vitro binder display platforms and their use in selecting for desired properties, including some aspects of developability, have been reviewed. 8In vivo methods for generating antibodies are also known, involving immunization of laboratory animals such as mice. Many approved antibody drug molecules have been discovered by using the animal's immune system to generate panels of antibodies, which are then screened in multiwell plates or by flow sorting for desirable properties (e.g., tumor-associated antigen binding, cytokine neutralization).

[0007] There is no guarantee that antibody genes isolated from "natural" sources, whether obtained directly from animals or incorporated into combinatorial display systems such as yeast display or phage display, will encode antibodies that exhibit good developability. This applies not only to antibody genes obtained from naive sources but also to those generated in vivo after immunization, and there is no a priori reason to assume that such antibodies will have good developability properties. The immune system aims to generate high-affinity antibodies, rather than to create antibodies that are useful for industrial manufacture or formulation as pharmaceuticals. Antibodies generated either in vivo or in vitro do not necessarily have the ability to be concentrated to levels suitable for pharmaceutical formulation. The average total concentration of IgG in human serum is 12 mg / ml, with any individual antibody being expressed at significantly lower concentrations. There is also great variability in the proportions of individual antibodies, with 1000-fold differences in expression observed between individual B cells during an immune response. 6 Therefore, regardless of the origin of the antibody (e.g., immunized animal / human donor, synthetic or semi-synthetic source), its solubility at high concentrations cannot be assumed.

[0008] Traditionally, the early stages of drug discovery focus on identifying molecules with desired therapeutic mechanisms of action, such as target-binding properties, while assessing developability is deferred until later stages, often after a promising molecule or a limited panel of molecules has been selected for preclinical development. An unfortunate consequence of this is that developability issues become apparent relatively late in drug discovery, by which time considerable time and money have already been spent. Such failures are costly. While the biopharmaceutical industry recognizes that candidate failure is a universal feature of drug discovery (the majority of promising drugs never reach the clinic), drugs that "fail early" are desired to reduce waste and allow resources to be diverted to rarer successes.

[0009] Engineering antibodies with a focus on improving affinity has been shown to generate affinity-enhancing variants with mutations that adversely affect biophysical properties, such as the tendency for self-association. One example is the anti-NGF antibody MEDI1912, which was affinity-matured from the parent antibody MEDI578. 7 It has been reported to have pM affinity for β-glucan. However, compared to MEDI578, the affinity-matured MEDI1912 antibody exhibited lower solubility, poor colloidal stability, aggregation at low concentrations, and a shorter half-life.

[0010] Efforts have been made to incorporate selection or screening of some aspects of developability into in vitro selection platforms such as phage display. For single domain antibodies or dAbs, which possess only the VH or VL domain, thermal challenge can reduce the proportion of antibodies with lower melting temperatures (Tm) prior to selection against antigen. An increase in Tm is associated with overall improvements in biophysical properties such as reversible unfolding, resistance to aggregation, solubility, bacterial expression and purification yield. However, molecules with similar Tm can exhibit differences in biophysical properties (Dobson et al., 2016). 7 and Example 3). ScFv molecules are also useful in bispecific platforms. 9~11Furthermore, phage display has been constructed that aims to introduce diversity into libraries while avoiding the inclusion of any potential post-translational modification sites (e.g., deamidation sites, isomerization sites, protease cleavage sites, and oxidation sites) (e.g., Tiller et al. (2013)). 12 ).

[0011] Numerous algorithms have been created for in silico prediction of molecular behavior, facilitating the comparison of the developability properties of a relatively large number of potential candidate drug molecules. This type of algorithm can help pinpoint the likely reasons underlying developability problems by identifying potentially causative amino acid sequence features. For example, it is recognized that hydrophobic patches in the sequence can cause the molecule to become "sticky," exhibiting nonspecific binding and / or self-aggregation, which are more likely to occur at high concentrations due to tightly associated polypeptides. Protein unfolding can expose hydrophobic residues that are buried within the molecule in its native structure.

[0012] When developability problems arise, whether anticipated or simply encountered during the development process, attempts can be made to address the difficulties using protein engineering. 13 identified specific positions in antibody VH and VL domains where the introduction of aspartic acid or glutamic acid residues improved biophysical properties. The resulting antibodies were said to be non-aggregating, well expressed, and capable of heat-induced refolding. The identified mutations reportedly increase aggregation resistance by altering the local charge distribution at specific positions, independent of the rest of the antibody sequence, and thus have been shown to serve as general templates for engineering human antibody variable domains with superior biophysical properties. In other cases, specific antibody sequences have been investigated for individual features that may limit their developability. For example, the anti-angiopoietin antibody "Ang2," identified from a B-cell hybridoma campaign, was found to have low levels of expression and be prone to aggregation in transient cultures. 6Analysis of the sequence identified an unpaired cysteine ​​at position 49 in the light chain variable domain framework. Twenty individual mutants were generated and evaluated for aggregation, and a solubility-improved version was identified in which cysteine ​​49 was changed to threonine (C49T). In this case, baseline expression of the parent clone was very low, and improved yield was observed in the C49T mutant without compromising binding.

[0013] Another example is a project that got stuck during development. 7 This is a "restoration" of the affinity-improved MEDI912 antibody described above. A combination of hydrogen:deuterium exchange and structural modeling identified three nonparatopic hydrophobic residues in the VH domain, which were reverted to those found in the parent antibody. Tryptophan at position 30, phenylalanine at position 31, and leucine at position 56 were converted to serine, threonine, and threonine, respectively (designated W30S, F31T, and L56T, where the number indicates the amino acid position within the antibody chain, the first letter indicates the original amino acid, and the last letter indicates the replacement amino acid). The improved developability version, designated MEDI1912-STT, exhibited reduced aggregation and was improved in numerous other aspects, including reduced nonspecific binding and increased half-life.

[0014] Bethea et al. (2012) 14 described an anti-IL-13 antibody with poor biophysical properties, including self-aggregation leading to precipitation at 13 mg / ml. An aromatic triad in the CDR3 of the heavy chain, consisting of phenylalanine, histidine, and tryptophan, was identified as a possible problem. The authors described the introduction of several mutations, including a single amino acid change (substitution of alanine for a tryptophan residue at position 100a) that improved solubility and reduced nonspecific interactions. In this case, this change also reduced target binding.

[0015] The example of MEDI912-STT and an anti-IL-13 antibody, where sequence optimization was able to simultaneously reduce both nonspecific binding and other undesirable behaviors such as self-aggregation, demonstrates that multiple developmental parameters may be interrelated and may have a common underlying cause. However, other cases have been reported in which nonspecific interactions were observed but there was little evidence of self-interaction. 2、15 .

[0016] Nonspecific interactions are an important consideration in drug discovery because they can adversely affect the performance, specificity, in vivo distribution, or half-life of a drug molecule. The in vivo half-life of an antibody can vary significantly among antibodies with the same Fc region. 16 , indicating the impact of antibody variable domains on half-life. This is often due to nonspecific interactions. If an administered drug is drawn into the "sink" of nonspecific binding interactions with non-target components, it may become less available for binding to its target molecule, reducing its ability to reach or penetrate the target tissue or lesion site. Even low-affinity nonspecific interactions can be significant, especially when the target is highly abundant. For example, circulating antibodies penetrate a vast area of ​​the endothelial glycocalyx (estimated area 350 m), which has been reported to reach a depth of more than 0.5 mm. 2 ) The negatively charged glycocalyx is composed of various glycosaminoglycans, such as hyaluronic acid, and proteoglycans, such as heparin sulfate, which together constitute the major components. It also presents absorbed plasma proteins. 17、18 Low affinity association with this matrix and other surfaces presented to circulating antibodies can have a profound effect on pharmacokinetics.

[0017] Methods exist for screening for nonspecific interactions, and these are typically performed on a clone-by-clone basis after individual binders of interest have been identified. Hotzel et al. (2012) describe a strategy that helps identify antibodies that exhibit nonspecific interactions by screening for binding to baculovirus particles in an ELISA. 16Xu et al. also described a multispecific reagent binding assay (PSR MFI) using a labeled protein mixture on a yeast display platform to identify antibodies with low specificity. 19 A number of chromatographic methods have also been developed to help identify antibodies prone to nonspecific interactions. These typically involve immobilizing the interaction partner, compound, or surface of interest on a matrix such as Sepharose, then passing test antibody molecules through the matrix and comparing the degree of retention between the test and control antibodies under fixed or varying wash conditions. Cross-Interaction Chromatography (CIC) 15 Methods such as chromatographic chromatography, hydrophobic chromatography, and heparin have been used.

[0018] An additional consideration for developability is the in vivo half-life of therapeutic proteins. To achieve optimal efficacy for many systemically administered drugs, their serum concentration must be maintained at a specific level (or within a target range) for a certain period of time. The half-life and / or effective in vivo concentration of a polypeptide can be influenced to some extent by the nonspecific binding interactions discussed above. Furthermore, for antibodies and other molecules containing an Fc region, half-life can be strongly influenced by binding of the Fc region to FcRn, a receptor expressed on endothelial cells. Human IgG has a relatively long half-life compared to other circulating molecules, which has been attributed to its pH-dependent interaction with FcRn. Following pinocytosis and endosomal transport, a relatively high-affinity interaction occurs between the antibody Fc receptor and the FcRn receptor, protecting the antibody from lysosomal degradation. At neutral pH, the affinity between Fc and FcRn is low, resulting in a neutral pH upon recycling to the cell surface. Under these conditions, the antibody is released back into the circulation. Therefore, pH-dependent binding to the FcRn receptor is an important property of IgG molecules. Crystal structure of FcRn complexed with rat IgG2a Fc. 20 By C H 2 and C H 3 Domains (C H 2 residues 252–254 and 309–311, and CH FcRn binding to the 3 residues 434-436 has been demonstrated and helps explain the important pH-dependent binding. Fc engineering has produced Fc variants with improved half-lives.

[0019] Suzuki et al. (2010) found a positive correlation between low affinity at neutral pH and long in vivo half-life, although other factors may clearly contribute. 21 The anti-IL-12 antibodies briakinumab and ustekinumab have half-lives of 8 and 22 days, respectively, despite their similar Fc domains. Using these antibodies and a series of cross-mutated versions, Schoch et al. (2015) 22 showed a good correlation between retained binding at neutral pH and in vivo half-life. Schoch et al. pointed to a large, positively charged patch on the VL of briakinumab as causing increased electrostatic interactions with FcRn, thereby limiting FcRn-IgG dissociation at extracellular pH. A similar pH-dependent interaction was described by Kelly et al. (2016), who argued that nonspecific interactions with other proteins may also contribute to the short half-life. 23 .

[0020] Typically, the primary goal in drug discovery using display libraries is to enrich a library for clones expressing binders with high affinity for binding to a target molecule of interest. Phage display libraries can be used to enrich binders from non-binders, enriching clones with higher affinity relative to clones with lower affinity. Affinity-based relative enrichment has enabled phage display to be used for affinity maturation of antibodies. Typically, biotinylated antigens are used to allow recovery of antibody:antigen complexes along with associated display packages (e.g., using streptavidin-coated magnetic beads). When the concentration of antigen is low, higher affinity antibodies in the population are more likely to form complexes than lower affinity antibodies, resulting in their relative enrichment.

[0021] However, when displaying binders in eukaryotic cells, approaches using limiting antigens combined with solid-phase capture on beads can be less effective than phage display at enriching for high-affinity binders. Unlike monovalent phage display, eukaryotic cells are polyvalent display packages carrying many copies of the same binder on their surface. The concentration of binders displayed within a cell population during selection can be relatively high compared to the antigen concentration and / or affinity being addressed. This can mask the effects of affinity, making it difficult to resolve differences in the affinities of binders displayed by different clones in a mixed population, and consequently reducing the enrichment factor achieved compared to approaches such as phage display.

[0022] Boder and Wittrup 24 described a method that claims to increase stringency and enable affinity selection in eukaryotic cell display libraries by using a limiting concentration of fluorescently labeled monovalent antigen in combination with flow cytometry to measure the amount of antibody bound per cell and control for differential expression. In this system, decreasing the concentration of target relative to binder reportedly increased binding stringency; therefore, higher affinity binders are said to be detected based on the level of signal detected from the target, since more molecules of target bound to the higher affinity binder than to the lower affinity binder.

[0023] Many publications describe the selection of monoclonal antibodies when target binding and antibody expression are simultaneously investigated. In this way, the degree of antigen binding can be normalized based on the level of display achieved. U.S. Patent No. 8,771,960 (DKFZ) describes the selection of higher affinity monoclonal antibodies using a fluorescence-activated cell sorting (FACS) method in which an antibody library or a group of different antigen-specific hybridoma cells is stained with PE-labeled antigen and counterstained with FITC-labeled protein G. Cells with the highest PE staining:FITC staining quotient were selected by FACS, and then individual cells producing antibodies with relatively high affinity for the target antigen used for selection were expanded. Therefore, the ratio of antibody-bound to antibody-unbound antigen was used as a direct measure of antibody affinity for that antigen.

[0024] Chao et al. (2006) 25 described genes encoding a repertoire of scFvs genetically fused to the yeast agglutinin Aga2p subunit. In the Aga2p yeast display system, binders of interest (here, scFvs) are fused to the Aga2p subunit, which binds to the Aga1p subunit present in the yeast cell wall via disulfide bonds. Yeast cells expressing target-specific binding molecules can be identified by flow cytometry using directly or indirectly labeled target molecules. For example, biotinylated targets are added to cells, and binding to scFvs displayed within the cell wall can be detected with streptavidin-phycoerythrin. Limiting the concentration of target molecules allows for enrichment of clones expressing higher affinity binders, as these clones capture more target molecules and consequently exhibit brighter fluorescence. To control for variations in scFv expression in different cells, Chao et al. (2006) 25used fluorescently labeled anti-tag antibodies to measure the antibody expression levels on the surface of each cell and normalize for variations in expression levels. This approach therefore made it possible to distinguish yeast cells displaying high-affinity binding molecules from cells expressing high levels of lower-affinity antibodies. Therefore, the purpose of measuring antibody display in this case was to normalize for differences in expression, thereby facilitating affinity selection.

[0025] A method has also been described in which the display level is used as an indicator of the potential expression yield of the same secreted protein. Aiming to identify highly expressing cell clones during the creation of stable cell lines for antibody production, International Publication No. WO2015128509 (Glenmark Pharmaceuticals) describes an approach in which antibodies are expressed in a secreted form, but a portion is "sampled" for display on the cell surface. This sampling occurs as a result of a splicing event that bypasses the first stop codon and splices the antibody gene to an exon encoding the transmembrane domain in a portion of the antibody mRNA. In this system, the antibody display level is reported to be directly correlated with the amount of soluble antibody expressed.

[0026] Many studies using yeast have reported a relationship between the level of surface display of binders on yeast cells, the thermal stability of the binders, and / or the yield of binders from cell culture. 26 fused soluble single-chain T-cell receptor (scTCR) variants to Aga2p and reported that the thermostability of the various variants strongly correlated with their soluble secretion levels and the amount of scTCR displayed as a fusion to Aga2p on the yeast cell wall. Shusta et al. proposed that intracellular proteolysis of thermodynamically unstable variants by the quality control machinery of the endoplasmic reticulum (ER) determines the efficiency of protein expression. Kowalski et al. 27、28investigated the soluble expression of mutant forms of the soluble polypeptide fibronectin type III (FnIII) ​​domain in yeast and also reported a correlation between the thermodynamic stability of the polypeptide and secretion efficiency (and therefore yield). Using a yeast display system, Hackel et al. 29 further investigated the effect of thermodynamic stability of the FnIII domain. Hackel et al. cultured yeast expressing a wide range of Aga2p-fused FnIII mutants to reduce their thermostability and found a positive correlation between thermostability and surface copy number. Thus, more thermolabile mutants exhibited reduced surface display levels.

[0027] International Publication No. WO 2012 / 158739 describes a two-step process for selecting polypeptides from libraries based on FnIII domains, involving antigen-based selection from an in vitro ribosome display library, followed by conversion of selected binders to a yeast display library for further antigen-based selection. While the in vitro ribosome display system produced mostly aggregated polypeptides, the inclusion of yeast display selection reduced the number of highly aggregated polypeptides in the selected population, although the proportion of clones producing monomers remained low. The slight improvement in the expression behavior of new clones was attributed to the yeast system's low tolerance for the expression of misfolded (e.g., thermolabile) proteins, and as a result, such proteins were less accessible for selection from the yeast library. Again, this work involved the Aga1p:Aga2p-FnIII display system.

[0028] On the other hand, Julian et al. 30found that simultaneous selection for antigen binding and surface display of VH domains on yeast resulted in antibodies with higher affinity but lower stability. The highest affinity VH domains were highly unstable. Because this study found only a small change (1.6-fold) in yeast display levels across the range of thermostabilities, the authors reported that the display level could not guide the selection of a set of mutants that improved both affinity and stability together.

[0029] Recent reports on the relationship between antibody affinity, specificity, stability, and solubility have described how improving one property (e.g., affinity) can lead to a decrease in another (e.g., stability), and how these trade-offs can be balanced to simultaneously optimize multiple antibody properties. 31 .

[0030] Therefore, extensive research in multiple fields has aimed to identify and understand the possible relationships between different properties of polypeptides that affect developability. Nevertheless, there remains a lack of polypeptide drug discovery methods that conveniently incorporate developability screening into the early stages of candidate drug selection, particularly for aspects such as drug solubility and avoidance of nonspecific binding. Summary of the Invention

[0031] The present invention provides methods and products that facilitate the detection of developability problems in polypeptides during the discovery stage from display libraries, thereby allowing molecules with adverse developability to be avoided and the pool of candidate drug molecules to be enriched for those with better developability properties.

[0032] The inventors surprisingly discovered that the level of polypeptide display on the surface of a eukaryotic host cell correlates with certain developability characteristics of the polypeptide, including its solubility in solution and its resistance to self-association. Host cells that express a polypeptide from a recombinant gene and display the polypeptide on their surface can be used to evaluate or screen such developability characteristics of a polypeptide by determining its level of display on the cell surface. This lends itself to parallel, high-throughput screening of multiple host cell clones, allowing comparison of relative surface display and selection of clones exhibiting polypeptides with better developability characteristics. Thus, the surface display level of a polypeptide is a predictor of developability in methods for screening polypeptide binders, such as antibody discovery. Furthermore, this correlation between the level of surface display and biophysical properties, such as self-association, allows binders with optimal biophysical properties to be selected or enriched within a binder library. Conversely, binders with poorer biophysical properties (e.g., lower solubility) can be selected for or excluded from a binder library.

[0033] The inventors have also devised methods for screening polypeptides expressed in higher eukaryotic cells in vitro for developmental aspects of the in vivo properties of polypeptide drugs, such as non-specific binding, half-life and effective concentration in serum or target organs and tissues.

[0034] The methods and uses of the present invention have particular advantages during early-stage screening, including screening of large and diverse libraries of binding agents such as antibodies. By incorporating developability screening of candidate polypeptide drugs at the earliest stages of drug discovery, the present invention reduces the risk of costly late-stage failure. Developability screening according to the present invention may also be used to select from later-stage pools of candidate polypeptide drugs, optionally from "families" of antibodies that share a common lineage, to enrich the pool for polypeptides with better developability and / or inform decisions about the selection of potential molecules for drug discovery. In addition, the techniques of the present invention may be used to identify improved variants of existing candidate polypeptide drugs, for example, drug candidates that fail to meet one or more developability criteria or for which improved developability characteristics are desired. Thus, methods for generating and rapidly screening derivative sequences exhibiting improved developability characteristics are described herein.

[0035] The polypeptide expression pathway in mammalian cells begins with the translation machinery (e.g., ribosomes) on the endoplasmic reticulum, followed by transport of the nascent polypeptide through the Golgi complex to the plasma membrane, where it is either secreted from the cell or retained at the cell surface (e.g., as a membrane protein). Once secreted, recombinantly produced polypeptides are quickly diluted into large volumes of culture medium and, after several days or weeks of accumulation, are typically present at concentrations of 1–100 μg / ml. This is low compared to the desired concentration of polypeptide drugs in pharmaceuticals formulated for patient administration. In contrast to secreted polypeptides, expressed polypeptides that are retained at the cell surface can form high local concentrations at the cell surface. Retention of expressed polypeptides at the cell surface significantly reduces the amount of polypeptide available, providing an opportunity to achieve high concentrations. Concentrations can be particularly high when the expressed polypeptide is expressed from a strong promoter, such as the cytomegalovirus (CMV) promoter. Therefore, when polypeptide binders are retained on the surface of mammalian and other eukaryotic cells in a recombinant cell library, the binders are enriched at high local densities on the cell membrane surface, especially when host cells are used that strongly express recombinant genes whose encoded polypeptides account for a significant proportion of total polypeptide synthesis. When applied to an entire panel of clones expressing a repertoire of polypeptides, inter-clonal variation in the surface display levels of different polypeptides can reflect polypeptide properties such as resistance to self-association. Polypeptides with a low tendency to self-associate can be enriched to higher levels on the cell surface, which is aided by the ability of the polypeptides to resist aggregation when brought into close proximity. Thus, eukaryotic cell display libraries in culture can serve as an in vitro selection environment for binders that exhibit good developability properties. This is supported by evidence in the examples presented herein.

[0036] According to a first aspect of the present invention, the surface display level of polypeptide binding agents (e.g., antibodies) on the surface of cultured eukaryotic cell clones is used as a predictor of the binder's developability properties, such as its solubility, resistance to self-association in solution, and / or ability to concentrate in solution. Without wishing to be bound by theory, one factor related to the solubility of a binder is its hydrophilicity; more hydrophilic (less hydrophobic) means higher solubility, greater resistance to self-association in aqueous solution, and higher concentrations that can be achieved in solution. Thus, the methods of the present invention can be used to distinguish more hydrophilic binding agents (candidate polypeptide drugs) from less hydrophilic binding agents based on their degree of surface display in the eukaryotic cell display system described herein.

[0037] The present invention provides providing a library of higher eukaryotic (e.g., mammalian) cell clones each containing DNA encoding a binding agent; culturing the clones in vitro under conditions for expression of the binding agent, wherein the binding agent is displayed on the cell surface; determining the surface display level of the binding agent on the plurality of clones; selecting one or more clones that exhibit higher surface display of the binder compared to other clones; and identifying binding agents encoded by one or more selected clones as having favorable developability properties.

[0038] The present invention may be used to distinguish or rank binders according to their developability properties and / or to select one or more binders with favorable developability properties. The developability properties assessed in such a method may be solution properties of the binder, such as solubility, resistance to self-association, and / or ability to be enriched in aqueous solution, as discussed in detail elsewhere herein.

[0039] Selection of clones with higher surface display results in a selected cell population enriched for clones with higher surface display of the binder, which may then optionally be used in one or more further methods, such as additional rounds of screening.

[0040] Methods for determining the surface display level are described in detail elsewhere herein and optionally include labeling the binder with an agent incorporating a detectable (e.g., fluorescent) label. For antibodies and other binders that contain an Fc region, it is convenient to label them with an agent that binds to the Fc region. For example, the detection agent may be a labeled anti-IgG antibody. The method may include determining the surface display level and observing the display levels of various binders on the cells of the library. Examples of copy number ranges are provided elsewhere herein.

[0041] A further aspect of the present invention relates to the evaluation of nonspecific binding during binder discovery in display libraries. It is advantageous to identify interactions with non-target molecules during initial binder discovery. The methods of the present invention may be used to screen a population of binders (e.g., antibodies) displayed on higher eukaryotic cells for optimal biophysical properties and a low tendency for non-specific interactions with non-target molecules in vivo. While it is conventional to select candidate polypeptide drugs for desirable binding to a target molecule (e.g., the molecular target of the polypeptide drug to which it binds in vivo and exerts a biological effect), problems with developability can also be mitigated if care is taken to negatively select against candidate polypeptide drugs that exhibit undesirable binding to non-target molecules (e.g., components or classes of molecules to which the binder exhibits non-specific binding, such as binding to negatively charged polymers such as nucleic acids). Non-target molecules can substitute for target molecules in binder selection methods, except that binders that recognize non-target molecules are discarded rather than retained, thereby enriching for binders that do not recognize non-target molecules.

[0042] The present invention provides methods for selecting binders with a low propensity for non-specific binding, enriching pools of drug candidates for those with low non-specific binding, and comparing the predicted pharmacokinetic performance of different drug candidate products. Such methods can be used during drug discovery, optionally in the early stages of screening, or to inform decisions about the selection of lead molecules for drug discovery.

[0043] The present invention provides (i) providing a library of eukaryotic clones each containing DNA encoding a binding agent; (ii) culturing the clones in vitro under conditions for expression of the binding agent, wherein the binding agent is displayed on the cell surface; (iii) exposing the binding agent to a matrix containing the non-target molecule to allow binding; (iv) discarding cells with higher levels of matrix attachment; (v) selecting cells with low levels of binding to the matrix and providing a population of selected cells enriched for clones expressing binders with low propensity to bind non-target molecules. The binders, and thereby the cells expressing them, are separated according to their relative binding to the matrix. Cells that bind to the matrix are discarded, and non-binding cells are collected.

[0044] High valency of antibodies displayed on the cell surface facilitates the detection of low-affinity cross-reactivity. When a population of cells displaying a binder is passed over a matrix, binding to the matrix can be evidenced by delayed passage. Cells displaying binders with a low likelihood of interaction can progress through the matrix more easily and be collected, in contrast to clones displaying binders exhibiting nonspecific interactions that emerge later. Thus, when a library of clones is passed through a matrix, clones displaying binders with a high propensity to bind to one or more components of the matrix will take longer to pass through the matrix or may even not emerge from the matrix at all, achieving clonal separation over time. The method can include collecting cells that pass through the matrix more quickly and discarding cells that pass more slowly and / or remain bound to the matrix. The matrix will generally comprise a solid or semi-solid substrate (e.g., beads, optionally packed in a column) to which one or more non-target components are immobilized. Several non-target molecules, such as heparin sulfate proteoglycans and other abundant components encountered in the bloodstream, can be tested. The matrix may comprise one or more components of the glycocalyx, e.g., hyaluronic acid, heparin sulfate. Alternatively, flow sorting or magnetic bead sorting may be used, with collection parameters based on the degree of binding to the desired interaction partner, compound, or surface for the non-target molecule. Thus, the present method may be used to enrich for cells expressing a binder that have a low propensity to bind non-target molecules in vivo in a mammalian subject to which the binder is administered.

[0045] The binder is (i) providing a library of eukaryotic clones each containing DNA encoding a binding agent; (ii) culturing the clones in vitro under conditions for expression of the binding agent, wherein the binding agent is displayed on the cell surface; (iii) exposing the binding agent to one or more non-target molecules to allow binding; (iv) discarding cells with higher levels of binding to one or more non-target molecules; (v) selecting cells with low levels of binding, providing a population of selected cells enriched for clones expressing binders with a low propensity to bind to the non-target molecule.

[0046] Such methods may be applied to cells (or samples of cells) of a library of higher eukaryotic cells displaying binding agents to enrich the library for cells expressing binding agents that have a low propensity to bind to one or more non-target molecules.

[0047] To facilitate the identification and / or separation of cells expressing a binder that recognizes a non-target molecule, the non-target molecule may be detectably labeled, for example, with a fluorescent label. Fluorescence allows cells to be separated by flow sorting in a FACS system, where unstained (unlabeled) cells can be distinguished from stained (fluorescently labeled) cells and sent to a collection or waste fraction accordingly. This may advantageously be combined with labeling to detect FcRn binding, and / or the presence and level of surface display of the binder (e.g., using an anti-Fc antibody to bind to a binder containing an Fc region), and / or target binding (e.g., using a labeled antigen). Using multiple different labels, e.g., fluorophores of different wavelengths, allows for simultaneous labeling and sorting for combined properties.

[0048] A further aspect of the present invention relates to screening polypeptides for pH-dependent interactions with the FcRn receptor. Antibodies (or other Fc-containing drugs) can interact with FcRn to extend or shorten half-life. As already mentioned, binding agents containing an Fc domain can interact with the FcRn receptor on endothelial cells and protect them from degradation. Operation of the FcRn recycling pathway is pH-dependent, requiring stronger binding within the low pH of the endosomal compartment to safely dock the polypeptide to the receptor, and lower affinity binding in the higher pH extracellular environment to release the polypeptide back into the bloodstream. In some cases, it is desirable to increase binding to FcRn for reasons beyond controlling half-life. For example, a "sweeping antibody" approach 32 It is desirable to use a nucleotide sequence that ensures that the administered antibody will engage well with FcRn at neutral pH and have a preferential interaction compared to other natural antibodies in serum. This antibody can then deliver the bound target molecule to the endosomal compartment, where it will be released by the decreased pH and subsequently degraded, for example, in the lysosome.

[0049] It may be desirable to combine selection of multiple aspects of developability (e.g., according to any such aspect of the invention described herein) or selection for target binding. To this end, selection is performed by exposing the binder to the target, allowing the target to be recognized by the cognate binder, thereby allowing clones displaying the cognate binder to become bound to the target. One or more clones displaying the cognate binder are then selected. The target may carry a detectable (e.g., fluorescent) label to facilitate selection of clones displaying the cognate binder. Advantageously, the method may include simultaneously determining the surface display level of the binder and the level of target binding by the binder, and simultaneously selecting clones displaying a cognate binder with higher surface display. Fluorescence-activated cell sorting (FACS) allows cells to be sorted according to their emitted fluorescence, and parallel selection for surface display and target binding can be performed by detecting surface-displayed versus bound targets using different fluorescent labels. Thus, clones that exhibit surface display above a selected threshold and also exhibit target binding can be selected.

[0050] A further aspect of the present invention relates to improvements in the selection of binders with high affinity for binding to a target of interest. The inventors have noted that while high levels of display of binders on cell surfaces can be advantageous in display libraries, this can undesirably limit the sensitivity or stringency of affinity-based selection for target binding. The inventors have recognized that limiting the display level of binders in display libraries facilitates the selection of binders with higher affinity.

[0051] According to this aspect, the present invention comprises: (a) providing an in vitro library of higher eukaryotic (e.g., mammalian) cell clones each containing DNA encoding a binding agent, wherein the encoded binding agent is expressed from a weakly active promoter and / or is expressed on the cell surface at a copy number ranging from 100 to 60,000 per cell; (b) exposing the library to the target and allowing recognition of the target by the cognate binder, thereby rendering the cells displaying the cognate binder bound to the target; (c) isolating the target-bound cells to provide a population of selected cells that display the cognate binder.

[0052] This results in a pool of clones enriched for clones that encode binding agents with higher affinity for the target. (d) exposing the selected population of cells to one or more rounds of selection on a target, optionally decreasing the concentration of the target to increase the stringency of the selection; and / or (e) selecting one or more clones that display a cognate binder with a desired level of binding to the target.

[0053] The concentration of target used may be predetermined or may be determined empirically by using a range of target concentrations and selecting an antigen concentration at which the degree of cell binding is higher than that seen with control cells (e.g., cells that do not express the binding agent or cells that express a binding agent that does not recognize the target).

[0054] Flow sorting may be used to identify a population of clones in a library that have the desired level of binding and / or binder display under different target concentration conditions. Selected clones may be identified based on the degree of cell-bound fluorescence. As the number of bound molecules decreases (due to a decrease in target concentration and / or a decrease in the level of binder display), the ability to distinguish between labeled and unlabeled cells decreases, especially when unlabeled cells exhibit a significant baseline of autofluorescence. In this case, using a retrievable target molecule (e.g., a biotinylated target) and magnetic beads (e.g., streptavidin-coated beads) to separate labeled cells may allow cells with significantly reduced levels of labeling to be separated from unlabeled cells, thereby allowing increased stringency to be used in the selection.

[0055] This method can be used to identify binders that recognize a target molecule with a desired affinity, to select binders according to affinity, and / or to enrich a pool of clones for clones that express binders with higher affinity for the target. The target can be any molecule of interest, such as a human receptor, ligand, enzyme, or other polypeptide.

[0056] The present invention provides libraries as defined in (a) above, and their use for the selection of binding agents with desired affinities for targets. Such libraries and methods for generating them are further described herein.

[0057] As outlined above, cell libraries can be used to distinguish binders based on different properties (developability and affinity) depending on the level at which the binder is expressed on the cell surface. During drug discovery, it may be desirable to select for both good affinity and good developability, in which case multiple aspects of the present invention may be combined. For example, cells that express binders at relatively low levels may be used to select for binders to a target of interest, resulting in a population of cells enriched for clones that display high-affinity binders. Their encoding DNA may then be provided to clones that express the binder at higher levels to select binders for the desired developability trait. These selections can be performed in either order (affinity selection followed by developability selection, or vice versa) and may include multiple rounds of selection (e.g., an initial affinity selection, then a developability selection, then an additional round of affinity selection).

[0058] By placing the surface expression of binders under an external control switch or regulatable element, dual-purpose display libraries can be constructed that are adaptable for use in both affinity and developability selection. For example, DNA encoding the binder may be operably linked to a promoter whose expression can be regulated by the addition of inducers or repressors to the cell culture medium. Controlling expression from outside the cell then allows the operator of the method to up- or down-regulate the level of expression at will. Thus, in the case of an inducible promoter, the promoter is activated by the addition of an inducer by an external operator, and thus can be considered to be externally inducible, even though induction ultimately occurs intracellularly. Several inducible promoter systems have been described, a typical example being the tetracycline-inducible promoter. 33 Dual-purpose libraries in which the expression of binder DNA is under external control (e.g., under the control of an inducible promoter) have the advantage that changing the expression level of the binder gene is rapid and simple, without the need to re-clone the encoding DNA into new cell populations.

[0059] Methods of binder display involving such libraries form part of the present invention. A method for identifying binders that recognize a target comprises: (i) providing a library of eukaryotic (e.g., mammalian) cell clones, each comprising DNA encoding a binding agent, wherein expression of the binding agent on the cell surface is externally regulatable (e.g., surface expression of the binding agent is under the control of an externally regulatable promoter) and the binding agent is displayed on the cell surface; (ii) culturing the cells of the library under conditions for low display on the cell surface (e.g., weak promoter activity); (iii) exposing the library to the target and allowing recognition of the target by the cognate binder, thereby rendering the cells exhibiting the cognate binder bound to the target; (iv) selecting cells that exhibit a cognate binder to provide a population of selected cells; (v) culturing the selected population of cells under conditions for increased presentation on the cell surface (e.g., greater activity of the promoter, optionally maximal activity); (vi) determining the surface display level of the binding agent on the plurality of clones, optionally by labeling the binding agent with an agent incorporating a detectable (e.g., fluorescent) label; (vii) selecting one or more clones that exhibit higher surface display of the binder compared to other clones.

[0060] Thus, by selecting / enriching for binders (and thus clones) with higher surface display, binders (and clones expressing them) with good developability properties are identified, selected, and / or enriched. Thus, the method can provide a pool of clones enriched for clones expressing binders with good developability properties.

[0061] Such a method effectively combines multiple aspects of the invention detailed elsewhere herein, namely, selection of a binder against a target and selection of cells based on surface display of the binder. Features of these aspects described elsewhere herein may be used in the combined method, including, for example, selection of the target concentration for stringent selection, the level of surface display of the binder, and the method of selecting the cells.

[0062] The regulatable promoter may be an inducible promoter. Depending on the type of inducible promoter system used, low-level or basal expression may be obtained in the absence of an inducer (e.g., a tetracycline such as tetracycline or doxycycline) in the medium. Alternatively, a low concentration of the inducer may be added. Expression from the promoter can be titrated by adding the inducer or by increasing the concentration of the inducer in the medium, preferably to obtain maximum promoter activity. An alternative to an inducible promoter is a repressible promoter, in which the initial state of the promoter is active and its activity is attenuated or blocked by adding a repressor to the medium.

[0063] It is often convenient to start with a library of basal expression states and perform one or more rounds of selection on the target before upregulating promoter activity to increase cell surface display of the binder and selecting for developability. However, in some cases, it may be desirable to first select for developability with the promoter activated, and then repress or allow the promoter's activity to fade (e.g., by removing an inducer from the medium) before performing affinity selection. This may be more convenient when using a repressible promoter. Thus, with reference to the numbered method steps above, steps (ii)-(iv) may be followed by steps (v)-(vii), or steps (v)-(vii) may be followed by steps (ii)-(iv).

[0064] The present invention further provides libraries for use in the above methods, examples of which are in vitro display libraries of eukaryotic (e.g., mammalian) cell clones containing DNA encoding a repertoire of binding agents, wherein expression of the binding agents (and hence their display on the cell surface) is under the control of a tetracycline-inducible promoter. Preferably, the encoding DNA is integrated at a defined locus in the cellular DNA. Such libraries can be providing a donor DNA molecule encoding a binding agent, and a eukaryotic (e.g., mammalian) cell; introducing donor DNA into a cell to create a recombinant cell in which the donor DNA has been integrated into the cellular DNA; creating a vector in which expression of the donor DNA is under the control of a tetracycline-inducible promoter for presentation at the cell surface; Culturing the recombinant cells to produce clones; Thereby, they may be produced by a method comprising providing a library of cell clones comprising donor DNA encoding a repertoire of binding agents.

[0065] Optionally, integration of the donor DNA is achieved by providing a site-specific nuclease within the cell, where the nuclease cleaves a recognition sequence in the cellular DNA to create an integration site at which the donor DNA is integrated into the cellular DNA, with integration occurring through DNA repair mechanisms within the cell.

[0066] Preferably, a tetracycline-inducible promoter is located on the donor DNA molecule encoding the binding agent, although it may be incorporated separately if desired. Generally, the donor DNA and / or cellular DNA of the recombinant cell will contain DNA encoding the binding agent downstream of a promoter for expression. Once the library is constructed, expression of the donor DNA can be induced from the promoter, and the cells can be cultured under conditions for displaying the binding agent on the cell surface.

[0067] Cells in which the expression of binder DNA is under the control of an inducible promoter also provide an opportunity to assess the rate at which cell surface display of the binder on expressing clones reaches a certain level and compare rates across multiple clones by comparing display levels after a short period of induction. This can be determined in advance (e.g., 4, 8, or 12 hours) or experimentally derived by observing the appearance of binder display after induction. Starting from the time expression from the inducible promoter is initiated, cell surface display can be plotted over time to observe the rate of increase in expression and surface display. Polypeptide expression will typically increase over time until it reaches a stable or equilibrium level at the cell surface. By determining and comparing cell surface display between clones early on, before the final expression level is reached, the first signs of developability aspects can already be seen. These developability aspects include the recoverable yield from expression as well as the ability of the polypeptide to be enriched in solution, its solubility, resistance to self-association in solution, and / or other developability characteristics discussed herein.

[0068] The rate of turnover, degradation, or internalization of a binder displayed on a higher eukaryotic cell can also be used as an indicator of such developability properties. This may optionally be assessed under conditions where the displayed binder is not continuously replenished with newly expressed binder (i.e., manifested as depletion of the binder from the cell surface, i.e., a reduced level of surface display). The binder may be labeled, unbound label washed away, and the rate at which the labeled binder is depleted from the cell surface over time due to degradation and / or internalization may be observed. Higher levels of binder depletion may be observed in some clones compared to other clones. Selecting clones with higher levels of surface display will select for clones with better developability properties (e.g., higher solubility, better resistance to self-association in solution, and less nonspecific binding).

[0069] In a library of cells, it may be observed that some clones exhibit a faster rate of increase in surface display than others, which can be used as an early indicator of the developability of the polypeptide, allowing clones with optimal properties to be selected from the library.

[0070] According to the method of the present invention, generally, after selecting a clone or a population of cells enriched with clones having desired properties, the selected clones or populations can then be used in one or more additional screening methods, examples of which are provided herein.The selected clones can be cultured together or individually.Methods can also be combined to allow clones to be screened for multiple properties, including target binding and various development properties as described herein.

[0071] The methods of the invention can be used to assess antibody behavior at high concentrations in the earliest stages of antibody discovery. In a preferred embodiment, the tendency of antibodies or other polypeptide binding agents to self-interact at high concentrations is detected by screening libraries of clones using library selection techniques such as flow sorting, bead-based selection, or chromatography.

[0072] After selection of one or more clones containing DNA encoding the desired binding agent in any embodiment of the invention, the nucleic acid encoding the binding agent may be recovered and / or the sequence of the nucleic acid encoding the binding agent may be determined. The nucleic acid (e.g., DNA) encoding the binding agent may be provided in an isolated form, for example, in a recombinant vector.

[0073] DNA encoding a binding agent of interest may be expressed in a host cell in vitro, optionally under conditions for expression of the binding agent in a soluble form. Thus, the host cell may secrete the binding agent, facilitating its recovery from the cell culture medium. The yield of the binding agent (e.g., from stably transfected host cells) may be, for example, at least 0.5 mg / ml, at least 1 mg / ml, at least 2 mg / ml, or at least 5 mg / ml. The binding agent may be purified and / or concentrated to provide an aqueous solution of the binding agent. Advantageously, the binding agent is provided in solution at a concentration of at least 1 mg / ml, optionally at least 10 mg / ml, at least 50 mg / ml, or at least 100 mg / ml. In some embodiments, the binding agent is provided in solution at a concentration of 50 mg / ml to 200 mg / ml, e.g., 50 mg / ml to 100 mg / ml.

[0074] Binding agents themselves identified or selected using the methods according to the invention are also provided herein as aspects of the invention, including those described in the Examples (including, but not limited to, binding agents comprising disclosed sequences, such as antibody VH and / or VL domain sequences set forth herein), and additional binding agents obtained as a result of practicing the methods of the invention. The binding agent of interest may be formulated into a composition comprising a pharmaceutically acceptable excipient. The invention extends to such compositions and their clinical uses, including the binding agent for use in a method of treatment of the human or animal body by therapy. The method may include administering the composition comprising the binding agent by subcutaneous administration. The composition comprising the binding agent in solution may be provided in a pre-filled syringe for injection, optionally in a kit containing one or more additional components, such as a needle and / or a product information leaflet containing instructions for administering the composition by injection, e.g., subcutaneous injection.

[0075] Eukaryotic cells in the context of the present invention are preferably higher eukaryotic cells, such as mammalian cells. Mammalian cells are commonly used for the large-scale expression of polypeptide pharmaceuticals (e.g., antibodies) intended for clinical use. As a result, it is advantageous to use mammalian cells when evaluating the properties of candidate polypeptides in drug discovery. Polypeptide binding agents can be expressed in their intended final molecular form during the early stages of discovery in mammalian cells. For example, if the desired clinical product is a full-length antibody (e.g., IgG), mammalian cells expressing full-length antibodies (e.g., IgG) can be used in the methods of the present invention.

[0076] The present invention can be most advantageously used in cell populations in which there is a fixed number (preferably one) of integrated binder genes per genome, to avoid copy number effects or heterogeneity resulting from differences in the number or identity of binder genes expressed in different clones in the population.In addition, the DNA encoding the binder is preferably integrated into the same locus in the cellular DNA of all clones, to avoid the complications of external influences on expression due to variations in the integration location of the coding DNA, which may otherwise arise from genome regions with different transcriptional activity.This has the benefit of transcriptional normalization of binder expression.Therefore, the present invention preferably uses multiple (e.g., large library) mammalian cell clones each containing DNA encoding a different binder sequence, where the coding DNA is located at a fixed locus in the cellular DNA and the encoded binder is displayed on the cell surface.Nevertheless, the benefits of the present invention can still be achieved using clones in which the coding DNA is randomly integrated into cellular DNA or provided on a plasmid for transient expression (e.g., Example 11). Integration of the DNA encoding the binding agent at a single or limited number of loci allows for better control of expression, if necessary, for example by using an inducible promoter, and preferably the DNA encoding the binding agent is present once per cell or once per chromosome copy in a diploid genome. Thus, it is preferred that the clones of the library each express only one or two members of the binding agent repertoire.

[0077] Various features of the present invention are further described below. Headings used throughout this specification are for navigational purposes only and should not be construed as limiting. Embodiments described in different sections may be combined where appropriate. DETAILED DESCRIPTION OF THE INVENTION

[0078] The headings and subheadings within this document are included solely to aid in searching and should not be construed as limiting. Multiple aspects and embodiments of the invention may be combined, and the methods for selecting polypeptide drugs based on developability will desirably encompass sequential and / or parallel combinations of the individual features and steps described herein.

[0079] Developability It is desirable to incorporate screening for multiple developability properties into the drug discovery process to provide insight into developability at an early stage and reduce developability risk. As detailed herein, binders (and their encoding cell clones) may be identified and selected for one or more desirable developability properties. Developability screening or selection in the present invention generally refers to predicted developability, where surrogate markers indicative of developability traits are assessed, allowing for high-throughput selection and the incorporation of developability selection into early-stage discovery. The present invention makes it possible to identify and reduce developability risk through enrichment of binders with more favorable developability properties. The developability property(ies) of individual polypeptides can then be directly confirmed by methods such as those described below. Thus, identifying a binder as having a particular property may include concluding that the binder is predicted to have that property based on data obtained from the methods of the present invention.

[0080] Identifying a binder as a candidate binder for development or identifying a binder as having good developability properties may include generating a report identifying the binder as having good developability properties. Identifying a pool of binders as being enriched for binders with good developability properties may include generating a report identifying the pool as being enriched for binders with good developability properties. Similarly, identifying selected cells as expressing a binder with good developability properties (or as being enriched for clones expressing the binder) may include generating a report identifying the cells (e.g., a selected cell population) as expressing a binder with good developability properties (or as being enriched for expression of the binder). The report may be a written report that may be provided in electronic format and / or in print. The report may provide quantitative and / or qualitative information regarding developability, for example, including data from one or more methods described herein.

[0081] Solubility, concentration, and self-association in solution Recombinantly expressed polypeptides usually need to be purified from cell culture and formulated at higher concentrations, e.g., for use as pharmaceuticals. For example, proteins may be expressed at 100 μg / ml in transient cell cultures or 1 mg / ml in stable cell cultures, providing the protein to patients at higher concentrations, e.g., 50 mg / ml or higher, upon subcutaneous administration.

[0082] Walker et al. (2008) 34 , Janson et al. (2012) 35Those skilled in the art will know of several simple techniques for expressing, purifying, and quantifying soluble binding agents such as antibodies. The concentration of a purified polypeptide binding agent in solution can be determined in several ways. For example, the absorbance of the solution at a wavelength of 280 nm can be measured, and this value can be used to determine the concentration based on the extinction coefficient of the polypeptide. Alternatively, a test sample can be compared with a standard protein of known concentration, and various colorimetric and fluorescent assays for this purpose are known to those skilled in the art.

[0083] A desirable development property is high solubility in aqueous solutions. The polypeptide may desirably be soluble at 10 mg / ml, 20 mg / ml, 50 mg / ml, 75 mg / ml, 100 mg / ml, 150 mg / ml, 200 mg / ml, or higher. The solution may be an aqueous buffer solution such as PBS. The solubility limit of the polypeptide may be greater than 10 mg / ml, 20 mg / ml, 50 mg / ml, 75 mg / ml, 100 mg / ml, 150 mg / ml, or 200 mg / ml. It is advantageous to provide the polypeptide in solution at a concentration well below the solubility limit to minimize molecular self-interactions and other undesirable effects that may occur as the solubility limit is approached further. Polypeptide self-association can lead to undesirable consequences in terms of product quality and stability, including increased viscosity or phase separation. Once the solubility limit of a molecule is reached, further increases in the dissolved concentration are no longer possible, resulting in undesirable effects such as precipitation of the molecule. In some embodiments, the solubility limit of the selected polypeptide binding agent (eg, improved variant) is between 10 mg / ml and 200 mg / ml, for example, between 50 mg / ml and 100 mg / ml.

[0084] After allowing a polypeptide solution to reach the point of precipitation and then filtering or centrifuging, it may be possible to determine the concentration of remaining soluble material to determine its solubility limit. This may also be referred to as the maximum solubility of the protein. However, there are more sensitive methods for measuring the onset of self-association, which may occur at concentrations lower than those required for precipitation. For example, at a critical concentration, monomeric molecules begin to form dimers and higher-order soluble aggregates. Such aggregates can be detected by various methods. The "critical concentration" is defined as the concentration at which self-interaction is evident using one or more of these methods. A desirable development characteristic of a binder is a high critical concentration, so that a pharmaceutical formulation of the binder in solution is well below the critical concentration. The critical concentration is preferably greater than 10 mg / ml, greater than 20 mg / ml, greater than 50 mg / ml, greater than 75 mg / ml, greater than 100 mg / ml, greater than 150 mg / ml, or greater than 200 mg / ml. In some embodiments, the critical concentration of the selected polypeptide binding agent (eg, improved variant) is between 10 mg / ml and 200 mg / ml, for example, between 50 mg / ml and 100 mg / ml.

[0085] Some consequences of self-interaction, such as precipitation, may occur over a long period of time, which is related to the shelf life of the product. Buffer composition, pH, and temperature may also have an effect. Therefore, parameters such as concentration can be determined under a set of reference conditions, for example, at 4°C using a standard buffer such as PBS at pH 7.4. The resistance of the polypeptide to self-interaction can be confirmed after a long period of time, for example, by testing after 4 weeks of storage under these conditions (and optionally at additional time points therebetween). Optionally, the polypeptide can resist self-interaction under such conditions for at least 6 months, confirming that the solution is below its critical concentration.

[0086] In this regard, poorly developable polypeptides are those that show obvious signs of self-aggregation even at lower concentrations in solution, e.g., 1 mg / ml or less in a standard buffer such as PBS at 4°C. An ideal polypeptide would have a critical concentration of 100 mg / ml or greater, i.e., one that resists such self-interactions and can be concentrated to 100 mg / ml. Between these extremes, improved polypeptides selected using the methods outlined herein may exhibit a critical concentration that is 1.5-fold or greater improved over the starting polypeptide. Thus, "improvement" can be defined with respect to the starting polypeptide, such as in the methods described elsewhere herein, where a variant is compared to a parent binder. Differences can be compared between different derivatives or variants of a parent binder, or more generally between binders from different clones, such as across a population of clones within a library, whether a naive library or an enriched population derived from another method (optionally a population of clones derived from the output of phage display selection or an antibody population derived from immunization). Using the present invention, clones encoding binders present at higher levels on cell surfaces may be compared with deselected clones encoding polypeptides at lower surface display levels within the same population. The biophysical behavior of polypeptides from selected clones can be compared with polypeptides derived from deselected clones from the same population at lower surface display levels. Antibodies or other binders that benefit from discovery using the present invention will be those that can be shown, by any method, to achieve higher concentration levels before the onset of self-interaction or that exhibit a reduced degree of self-interaction within a given assay compared to the comparison polypeptide. For example, a starting polypeptide that shows evidence of self-aggregation at 10 mg / ml can be improved to 15 mg / ml or greater. Alternatively, the optimal polypeptide selected from a library may resist self-aggregation at 10 mg / ml, while other polypeptides derived from rejected clones within the same population may exhibit self-interaction at 10 mg / ml or less. In either case, the parent polypeptide or the polypeptide derived from the rejected clone is referred to as the "comparison polypeptide."

[0087] As a control for the self-association and other biophysical properties selected in the present invention, an antibody with known desirable properties may be used to compare the relative performance of the parent antibody and its improved derivatives (or selected library members versus deselected members). The National Institute of Standards and Technology uses the NIST RM 8671 antibody as a "gold standard" reference antibody, which is certified by the American Chemical Society. 36 The antibody has been extensively characterized by over 100 collaborators using comparisons published as part of a 3-volume series by NIST. The NIST RM 8671 antibody was shown to have minimal self-interactions (Saro D et al, Developability Assessment of a proposed NIST monoclonal antibody 37 Alternatively, adalimumab antibodies have been shown to have minimal self- and cross-interactions, and several studies have shown this, e.g., Jain et al. (2017) 2 and Sun et al. (2013) 38 The critical concentration or solubility of a binding agent, particularly an antibody, described herein may be compared to one or more such reference antibodies for benchmarking purposes.

[0088] Self-interaction can be determined in several ways, allowing for direct and quantifiable comparisons between antibodies selected for high levels of surface display compared to antibodies deselected based on lower display levels. Size exclusion chromatography (SEC), e.g., high-pressure liquid chromatography (HPLC), allows for the separation of monomeric and multimeric forms of polypeptides (including dimers and higher-order forms). The proportion of dimeric and / or multimeric material can be quantified, and the degree of multimer formation can be compared between binders (e.g., between the parent binder and the improved variant), which can be detected as an increased retention time and / or a broader elution peak after column passage. HPLC-SEC profiles can be determined at a single concentration, and the proportion of multimers and / or retention times are compared. A detectable and reproducible decrease in either retention time or peak width would be considered an improvement. A reduction in the multimer peak to 60% of its value in the parent clone is considered an improvement. Alternatively, the critical concentration at which a threshold for multimerization occurs can be determined for each clone by testing a range of concentrations and determining the concentration at which this threshold for multimerization occurs (e.g., 5%). Alternatively, the increase in multimerization can be plotted over time and the rate of multimer accumulation compared. 2 In either case, an increase in the critical concentration represents an improvement. The magnitude of the improvement in the critical concentration may be, for example, at least 1.5-fold, or at least 2-fold.

[0089] The improvement in solubility limit, or critical concentration, is optionally 1.5-fold to 50-fold, such as 1.5-fold to 15-fold, 1.5-fold to 10-fold, or 2-fold to 10-fold.

[0090] Methods for determining polypeptide self-association include self-interaction chromatography (SIC). In this technique, a binding agent, such as an antibody, is immobilized on a matrix and a solution of the same binding agent is passed over the matrix. Extended retention times compared to a control antibody indicate self-interaction. 38Alternatively, a high-throughput approach can be used, in which different binders are immobilized on a biolayer interferometry chip (BLI) and tested for self-interaction by immersing a solution of their soluble form, yielding a signal related to the amount of soluble binder that binds. This "Antibody Clone Self-Interaction using Biolayer Interferometry" (CSI-BLI) approach has been shown to correlate well with more laborious approaches such as SIC and requires less material. In both cases, the retention time (for SIC) and the signal achieved (for CSI-BLI) of the test binder can be compared to the parent binder and a control binder known to resist self-aggregation (such as the benchmark antibody mentioned above).

[0091] In affinity capture self-interaction nanoparticle spectroscopy (AC-SINS), test antibodies are displayed on cell surfaces to assess their potential for strong self-interaction with other bead-displayed antibodies. 39~41 The resulting decrease in interparticle distance can be detected as an increase in the plasmon wavelength of the colloidal gold solution. AC-SINS is a preferred technique for determining the critical concentration because it is a sensitive and simple assay. In a preferred embodiment, the improvement in the developability of a binder can be detected as an increase in the critical concentration of the polypeptide in solution, which can be measured by measuring the change in plasmon wavelength in an AC-SINS assay.

[0092] Self-association was characterized by static and dynamic light scattering (DLS). 42、43 DLS can also be used to measure the size and shape of molecules in solution, allowing the hydrodynamic radius and percent polydispersity of the molecules in the sample to be calculated. In DLS, the mutual diffusion coefficient is evaluated in relation to antibody concentration as a measure of self-association. Analytical ultracentrifugation 44 , membrane osmolality measurement 45 , and neutron scattering 46Other methods of measuring self-interactions can be used, such as

[0093] The methods of the invention comprise selecting one or more clones that have increased surface display of a binding agent relative to other clones, and identifying binding agents encoded by the one or more selected clones as having good solubility in solution and / or resistance to self-association, wherein the critical concentration of the binding agent(s) is increased by at least 10%, at least 25%, or at least 50% relative to binding agents encoded by one or more other clones in the population, or relative to a parent binding agent of which the selected binding agent is a mutant form, and / or the critical concentration of the binding agent(s) is increased by at least 1.5-fold, at least 2-fold, at least 3-fold, at least 5-fold, at least 10-fold, at least 20-fold, or at least 100-fold relative to binding agents encoded by one or more other clones in the population, or relative to a parent binding agent of which the selected binding agent is a mutant form.

[0094] As discussed, the methods herein can be used to filter out (or reduce the frequency or prevalence of) binders with poor developability and encourage the selection of binders with better developability properties (by increasing their frequency or prevalence, enriching for them). Such methods go beyond simply identifying and avoiding "problem" binders. The sensitivity of the techniques described herein allows them to be used to distinguish the best (e.g., most soluble) candidates among multiple binders with good solubility. The present invention can also be used to exert selective pressure that favors maintained or enhanced developability during affinity-based selection.

[0095] The comparison polypeptide can be any binder from an unselected clonal population (or a clone whose binder display level is below a determined threshold level). The comparison polypeptide can have a solubility limit of at least 5 mg / ml, at least 10 mg / ml, at least 20 mg / ml, at least 30 mg / ml, at least 40 mg / ml, or at least 50 mg / ml, e.g., as determined by any method described herein. The comparison polypeptide can optionally have a critical concentration of at least 5 mg / ml, at least 10 mg / ml, at least 20 mg / ml, at least 30 mg / ml, at least 40 mg / ml, or at least 50 mg / ml, e.g., as determined by any method described herein. A binder from a selected clone may exhibit at least a 10%, at least a 20%, at least a 30%, at least a 40%, at least a 50%, at least a 60%, at least a 70%, at least a 80%, at least a 90%, or at least a 100% improvement in a developability characteristic, e.g., a solubility limit or a critical concentration, compared to a comparison polypeptide (e.g., a clone from an unselected population, or a parent binder). A binder from a selected clone may exhibit at least a 1.5-fold improvement in a developability characteristic, e.g., an increase in a solubility limit or an increase in a critical concentration, compared to a comparison polypeptide. The comparison polypeptide may optionally have a critical concentration of at least 10 mg / ml, at least a 20 mg / ml, at least a 30 mg / ml, at least a 40 mg / ml, or at least a 50 mg / ml. When comparing the properties of a polypeptide to a comparison polypeptide, it will be understood that the comparison is performed under conditions identical to those standard in the art for use as a control.

[0096] This method may include, for example, experimentally measuring the critical concentration by determining the change in plasmon wavelength in an AC-SINS assay to confirm the improvement of the critical concentration. Standard buffer conditions for determining parameters such as solubility limit and critical concentration include PBS at pH 7.4 at 4°C. The parameters may be measured immediately after the newly synthesized polypeptide is formulated into the aqueous buffer solution. Alternatively, the parameters may be measured after a storage period, for example, 1 hour, 1 week, 2 weeks, 28 days, or 6 months under standard buffer conditions.

[0097] It may be desired to determine the parameter after exposure to elevated temperatures, in which case the temperature may be increased to, for example, 25°C, 37°C, or 50°C during storage before returning the solution to 4°C for measurement.

[0098] non-specific binding Nonspecific binding, "multireactivity," "multispecificity," or "low specificity" refers to the interaction of a binder with multiple non-target molecules in addition to its cognate target. For example, a polypeptide may nonspecifically bind to hydrophobic, negatively charged, or positively charged surfaces. Multireactive polypeptides may be described as exhibiting "stickiness," reflecting their binding to non-target molecules through this type of interaction in addition to the specific recognition between the binder polypeptide and its target. Nonspecific binding often manifests as a low-affinity interaction, but can still be problematic when non-target molecules are abundant. Multireactivity may result from clustered hydrophobic or charged amino acid residues on the surface of the binder polypeptide (in the case of antibodies, this may be in the heavy and / or light variable domains), leading to a class of non-specific interactions with other (non-target) molecules. For example, a polypeptide may exhibit binding to hydrophobic surfaces, which can occur if the polypeptide displays one or more hydrophobic patches on its surface or if hydrophobic patches are exposed by unfolding of the polypeptide in solution. Alternatively, polypeptides may exhibit binding to molecules that carry a net negative charge (at neutral pH), such as negatively charged surfaces or the negatively charged backbone of DNA, or other negatively charged polymers, such as heparin or heparan sulfate. Regardless of the underlying molecular motif responsible for these nonspecific interactions, multireactivity is generally an undesirable characteristic for candidate polypeptide drugs. Multireactivity of polypeptide drugs is associated with poor pharmacokinetics, such as short half-life in vivo, and / or poor uptake of the drug into tissues from the circulation.

[0099] In embodiments of the invention, cells expressing a binding agent (e.g., a library, or a sample of cells from a library) are exposed to one or more non-target molecules or "multireactive probes." This can be used to distinguish cell clones within a population that display binding agents that interact with the multireactive probe from cell clones that display no or little binding to such probes. The multireactive probes may include any one or more of the following: nucleic acids (e.g., DNA), streptavidin, heparin, heparan sulfate, chondroitin sulfate, carboxyl dextran, other sulfated proteoglycans, insulin, lipopolysaccharide, baculovirus, KLH, FcRn, laminin, collagen, trigger factor, Hsp70, Hsp90, or other heat shock or chaperone proteins, hyaluronic acid, or other glycocalyx components. The non-target molecules may be displayed in isolated form or as a mixed preparation; for example, membrane preparations from mammalian cells (e.g., CHO cells) may be used to test for multireactivity. The non-target molecule may be one found in vivo, e.g., in a mammal, e.g., a human. It may be found in the extracellular matrix or bloodstream and / or on the surface of a cell. Synthetic surrogates of the non-target molecule may be prepared and used as multireactive probes.

[0100] Multireactive probes may be selected to screen for specific modes of nonspecific binding, for example, to detect nonspecific binding to negatively charged surfaces. Multireactive probes may be selected that carry a net negative charge at neutral pH, such as molecules bearing negatively charged surface regions, such as heparin sulfate, DNA, or streptavidin, and / or FcRn. Suitable multireactive probes may be identified based on their calculated isoelectric point (pI). pI is a measure of the charge of a protein and is defined as the pH at which the protein carries no net charge. To detect binders that exhibit nonspecific binding to negatively charged surfaces, multireactive probes with a pI of less than 6 may be selected. Examples are known to those skilled in the art and include streptavidin, nucleic acids, and sulfated proteoglycans such as heparin sulfate or carboxyl dextran. Conversely, to detect nonspecific binding to positively charged surfaces, multireactive probes with a positively charged surface region and / or a base pI (e.g., a pI greater than 8) may be selected. Such probes may also be used to detect binders with negatively charged surface patches that may result in undesirable electrostatic repulsion from negatively charged cell surfaces in vivo. To probe multireactivity via hydrophobic attraction, multireactive probes with one or more hydrophobic regions on their surface, such as Hsp70 or Hsp90, may be selected.

[0101] Several methods are available for screening individual clones for low specificity: Cross-Interaction Chromatography (CIC) 47 is a method in which a target molecule or mixture, often a polyclonal antibody preparation, is immobilized on a chromatography matrix and the retention times of the various antibodies are measured. Delayed retention due to interactions with the immobilized molecule or the resin itself indicates poor specificity. This approach has been used in several studies to characterize recombinant antibodies (e.g., 2、15、38). As described above, size exclusion chromatography may be used to determine the interaction of the binder with the matrix. In this context, it will be understood that the non-target molecule(s) of interest occupy the place of "target" in these methods, being the molecular component with which the binder interaction is being tested.

[0102] Characterization of antibodies and other polypeptides has also been performed using alternative matrices or interactions with target molecules. For example, cells may be screened for their binding to abundant molecules present in the glycocalyx, e.g., heparan sulfate proteoglycans (HSPGs), which consist of a core protein to which heparan sulfate (HS) glycosaminoglycan (GAG) chains are attached. Given the large surface area and abundance of such molecules in the glycocalyx, this is a particularly suitable class of molecules to test. Heparin affinity chromatography involves passing a sample through a matrix containing immobilized heparin. 48 Ready-made resins are available from several sources (Heparin Sepharose (Pharmacia), Bio-Gel Heparin (Bio-Rad, Vienna, Austria), Eupergit Heparin (Riihm Pharma, Weiterstadt, Germany), and Toyopearl Heparin 650M (TosoHaas, Stuttgart, Germany)). Bound materials will be retained or their passage will be retarded. Unbound materials will pass through or be removed during washing. Bound materials are removed using altered buffer conditions, e.g., increasing salt concentration. This method can be used to screen soluble antibodies and determine the extent to which they interact with heparin. Hydrophobic interaction chromatography can also be used to characterize antibodies by their tendency to interact with a hydrophobic matrix. 49、50 . When one or more non-target molecules are presented on a matrix, binding of the binder to the matrix manifests as increased binding of cells to the matrix, e.g., a slower passage of cells through the matrix, or binding to beads (e.g., magnetic beads) coated with the non-target molecule(s). Thus, binders with greater binding to one or more non-target molecules can be isolated by removing (discarding or not selecting) the fraction of cells with greater binding, while cells with less binding can be collected and, optionally, selected for further steps. As described above, comparison and control polypeptides can be used to confirm improvements resulting from the present invention. Improvements can be determined by identifying concentrations at which non-specific interactions are evident, optionally defined under standard conditions. Alternatively, the extent of non-specific interactions can be determined at a defined concentration. Methods for assaying non-specific binding generate a "stickiness measure" of the binder, providing a quantitative measure of non-specific binding that can be compared with other binders. Thus, the "stickiness measure" of any given assay is the difference between the values ​​for the test polypeptide compared to a control polypeptide known to have very low non-specific interactions. For example, the approved antibody adalimumab 2、38 Alternatively, antibody NIST RM 8671 exhibits minimal self-interaction or association with other polyclonal IgG molecules, as estimated by self-interaction chromatography (SIC) and cross-interaction chromatography (CIC) methods. (Saro D et al, Developability Assessment of a proposed NIST monoclonal antibody 37 ). Approved polypeptides according to the present invention are those that exhibit an observed improvement in the "stickiness measure" compared to a comparison clone. The comparison clone may be the starting clone that has been improved, or may be a clone that has been deselected by use of the present invention.

[0103] Some methods, including these chromatographic methods, are typically used to characterize individual antibodies and generate "adhesion assays." Chromatographic matrices can be used to separate cells based on their interaction with the immobilized molecules. For example, lectins immobilized on cyanogen bromide-activated Sepharose have been used to separate T cell populations. 51 Such systems can be modified to separate antibody-expressing cells based on their interaction with an immobilized target, such as a polyclonal antibody, heparin sulfate, or other test molecule. If there is an interaction with the immobilized target or support matrix, cells bearing such antibodies will be retained or delayed relative to non-interacting cells. When used to separate cells displaying antibodies with different binding propensities, loading or washing buffers can be modified to achieve the desired stringency.

[0104] Non-chromatographic methods can also be used to identify and quantify low specificity within individual antibodies. For example, Hotzel et al. (2012) use antibody binding to baculovirus particles in an ELISA to identify antibodies that exhibit non-specific interactions. 16 In a similar manner, other test molecules (e.g., heparin sulfate) can be immobilized or displayed on beads and tested for interaction with individual antibodies. Non-chromatographic methods such as these can be adapted to identify clones exhibiting low specificity from libraries displayed in higher eukaryotes.

[0105] A mixture of detergent-solubilized membrane proteins has been prepared, biotinylated, and used to identify clones within a yeast library that display less specific antibodies. 19While the presence of detergents may be tolerated using yeast libraries, it may not be suitable for display systems based on higher eukaryotes, such as mammalian cells. Test molecules or mixtures used to identify less specific interactions are labeled with molecules such as fluorophores or biotin, facilitating labeling and recovery of the molecule and its complexes on streptavidin-coated surfaces. For example, molecules such as fluorophore-labeled chondroitin sulfate or heparin sulfate (e.g., those from AMS catalog numbers AMS.CSR.FACS-A1, C1, or D1, or E1, or AMS.CSR.FAHS-P1) can be used to separate clones in flow sorting according to the degree of interaction with the labeled test molecule. High-affinity expression of polypeptides on the cell surface increases the sensitivity of this approach, especially when multivalent target molecules are used. Clones within the library can be separated by flow cytometry based on the binding of fluorescent molecules. These test molecules can be used in conjunction with other labeled molecules to pre-, simultaneously, or subsequently select for other desirable properties, such as target binding, or other binding / evasion of other molecules of interest, such as Fc receptors. Other methods include AC-SINS, as described above. In this technique, test binding agents are displayed on the cell surface and their potential interaction with one or more non-target components displayed on other cells or beads is assessed. 39 The resulting decrease in interparticle distance can be detected as an increase in the plasmon wavelength of the gold colloid solution.

[0106] Cells expressing a binder may be exposed to one or more non-target molecules where a mixture of clones (e.g., a library or a sample therefrom) is pooled in one container, which is convenient for methods such as FACS, or chromatographic techniques. In other cases, cells expressing a binder may be exposed to one or more non-target molecules in separate containers, e.g., one clone per container, and then the resulting levels of interaction or binding may be measured and compared individually, which may be more convenient for methods that measure interparticle distances, such as AC-SINS, or when comparing relatively small numbers of binder-expressing clones.

[0107] Thus, a fluorescently labeled multireactive probe can be mixed with a population of cells, and a detection or separation method is used to distinguish cellular clones that express the multireactive binder (identified by binding to the multireactive probe) from clones that express binders that do not. Cell clones that cannot bind to the labeled probe can be separated by flow sorting, magnetic bead separation, and other separation methods to enrich for clones that express the multireactive antibody. Example 6 demonstrates that not only is it possible to achieve sufficient differentiation between multireactive and non-multireactive clones within a population, but that this can be done using simple and practical steps that allow for their separation.

[0108] Identifying drug-targetable mutations While the present invention can be used at all stages of drug discovery, including early-stage selection of binders (e.g., from naive libraries or selected populations obtained from immunization or other display approaches), and later to compare the quality of shortlisted panels of candidate molecules, it also finds use in situations where a binder of interest has already been identified but is later found to require improvement in one or more developability properties. Binders identified from any source may be found to exhibit less-than-ideal developability properties, and in such cases it may be preferable to improve the sequence of an existing molecule rather than starting a new drug discovery program from scratch to find an alternative molecule.

[0109] The methods of the invention may be used to identify variants of a binding agent, where the binding agent has been identified as needing improvement in one or more developability properties (e.g., self-association, solubility, nonspecific binding, and / or other contemplated properties), and the invention is used to predict whether one or more variants will exhibit improved developability. Thus, the selection methods of the invention may be performed on a population of cells displaying variants of a "parent" binding agent. These may be referred to as libraries, and in some cases will display a large and diverse population of variant binding agents, although in some cases the number of clones compared may be relatively small, e.g., up to 10. Thus, the methods of the invention may include providing a library or a plurality of clones, where the binding agent is displayed on the cell surface, and the clones are produced by introducing generated variant sequences of the parent binding agent sequence and DNA encoding the variants into cells, allowing the DNA to integrate into the cellular DNA. Suitable methods and techniques are described in detail in other sections of this document.

[0110] The "parent" binding agent from which the variants are generated may be one identified as needing improvement due to poor performance in one or more developability assays. Alternatively, it may simply be desirable to investigate whether its developability can be improved by sequence changes. Various aspects of developability are discussed herein, and the parent molecule may be identified as needing improvement in any of these. The parent molecule may be found to have a solubility limit (maximum solubility) of less than 50 mg / ml, less than 20 mg / ml, less than 10 mg / ml, less than 5 mg / ml, or less than 1 mg / ml. The parent molecule may be found to have a critical concentration of less than 50 mg / ml, less than 20 mg / ml, less than 10 mg / ml, less than 5 mg / ml, or less than 1 mg / ml. The parent may exhibit undesirable aggregation in solution and / or may not be able to be concentrated above 1 mg / ml in solution without aggregation and / or precipitation. The parent may exhibit nonspecific binding to one or more non-target molecules. The parent may be identified as needing improved binding to and / or dissociation from FcRn.

[0111] Optionally, a bioinformatics evaluation of the parent polypeptide sequence is performed to identify possible sequence features that are predicted to be mutated to improve performance by mitigating identified developability issues. Thus, one or more amino acid positions predicted to be associated with developability (e.g., solubility, self-association, non-specific binding) can be identified.

[0112] Such bioinformatics evaluation can be used to inform mutation strategies. Thus, variants of the parent polypeptide sequence can be generated, optionally including mutations at one or more amino acid positions identified in the bioinformatics evaluation. Thus, mutations can be made at one or more amino acid residues in the parent binding agent polypeptide sequence that are predicted to promote self-association, aggregation, and / or nonspecific binding, and / or reduce solubility. The mutations generate DNA encoding one or more variants of the parent sequence, which can be introduced into higher eukaryotic cells to generate a population of cells encoding the variant binding agents (methods for which are described herein). The cells can be cultured under conditions for expression of the binding agent, whereby the binding agent is displayed on the cell surface. A plurality of cells expressing the binding agent can be used as a library, as described herein, and selection can be performed to identify clones with higher surface display of the binding agent as an indicator of improved developability characteristics.

[0113] In various examples described herein, sequence analysis is used to identify potentially problematic residues. Alternatively, random sequence variation can be used. Methods for generating mutant and derivative libraries are described elsewhere herein. Individual mutants can be made and evaluated for improved biophysical properties. The ability to create large libraries of many such mutants and directly select for improved properties, such as resistance to self-aggregation, greatly facilitates the discovery of antibodies and other binding agents with optimal solubility properties, particularly when paratopic residues are involved, as changes that benefit solubility may also diminish target binding. 14 The methods of the invention may combine selection for developability with selection for retained target binding. Such selections are optionally performed simultaneously, and methods for simultaneous or sequential screening for affinity and solubility are described.

[0114] As discussed, methods of the invention that involve selection based on the level of surface display of a binding agent may be used to identify variants with improved solution properties. For example, methods that improve the developability properties of a "parent" binding agent (e.g., an antibody) include: introducing mutations into the amino acid sequence of the binding agent to generate variants; introducing DNA encoding the variant into eukaryotic (e.g., mammalian) cells to provide a plurality of cell clones each containing DNA encoding the derivative antibody; introducing DNA encoding the parent binding agent into eukaryotic (e.g., mammalian) cells to provide cell clones containing DNA encoding the parent; culturing the clones in vitro under conditions for presenting the binding agent on the cell surface; determining the surface display level of the binding agent in the plurality of clones; selecting one or more clones that exhibit increased surface display of the derivative antibody compared to the parent-expressing clone; and identifying one or more variant binding agents encoded by one or more selected clones as having improved developability properties compared to the parent antibody.

[0115] There is a relationship between the propensity for self-interactions and the propensity for non-target interactions. It has been shown that a small number of amino acid changes (1-3) can have beneficial effects on both aspects. 6、7、14 In other cases, self-interactions may be limited, showing evidence of low specificity. In either case, methods of the invention, including the selection of binders with less non-specific binding, may also be used. Thus, as discussed elsewhere herein, the present system may be used to identify undesired non-specific interactions with other molecules, also referred to as "low specificity" and "multispecificity." The invention has the advantage of conducting such screening in the context of expression on higher eukaryotic cells, such as mammalian cells, with modifications such as glycosylation that more closely reflect those found in production cell lines typically used to manufacture clinically used products. High display levels of polypeptides on cells 23This would serve to increase the sensitivity of the system in detecting low-affinity, undesired interactions by increasing the affinity of any interactions. Furthermore, the surface of higher eukaryotes themselves (or the environment of the endoplasmic reticulum and Golgi apparatus) can serve as a matrix that exposes binders to diverse polypeptides at relatively high concentrations, allowing less specific binders to interact with non-target molecules on the same or adjacent molecules. This results in reduced levels of presentation. This may also cause the presenting cells to aggregate with other cells in the population. The resulting cell aggregates can be removed (e.g., by filtration or sedimentation) to deplete such clones from the population. Removal of aggregated cells can also be used to reduce the representation of self-interacting clones.

[0116] In a further application, host cells expressing undesired targets from endogenous or exogenous genes can be used to deplete cross-reactive clones. For example, endothelial cells expressing components of the glycocalyx or mammalian cells transfected with the gene of interest. Clones that bind to the target may be depleted based on low surface expression or cell aggregation.

[0117] Biophysical measurements at a single concentration of polypeptide may be compared between the starting clones and improved clones generated using the present invention. Alternatively, concentrations at which undesired biophysical parameters are measured may be compared. The method may involve selecting variants with improved one or more desired developability properties, such as higher solubility, lower tendency for self-association in solution, lower nonspecific binding to non-target molecules, higher critical concentration, etc. Fold differences or % differences may be measured, and example values ​​are provided elsewhere herein. Comparisons will typically be made to the parent binder. However, comparisons of variant polypeptides from clones selected according to the present invention with comparison polypeptides deselected according to the present invention may also be used to confirm and quantify improvements when the present invention is used to select for other biophysical properties described herein.

[0118] Methods for generating mutant and derivative libraries According to any of the selection methods discussed herein, DNA encoding the displayed binding agent can be recovered from one or more selected cells, optionally mutated to generate variants, and / or subcloned to allow for additional rounds of selection, for example, in a second display system. This may be another eukaryotic display system that provides a different level of surface display, or may be an entirely different display system, such as phage display. The second system may use secreted expression and / or may allow for direct functional selection, as previously described (see references). 52~54 and International Publication No. WO2015 / 166272. Alternatively, the input DNA encoding the binding agents can be a population of binding agents from an unselected library, or a population derived from immunization or another display technology such as yeast or phage display.

[0119] Thus, following generation of a library by the methods of the invention, one or more library clones may be selected and used to generate additional, second-generation libraries. If the library is generated by introducing DNA into eukaryotic cells as described herein, the library may be cultured to express the binders, and one or more clones expressing the binder of interest may be recovered, e.g., by selecting for binders against a target as described elsewhere herein. These clones may then be used to generate derivative libraries containing DNA encoding a second repertoire of binders.

[0120] In other instances, it may be desirable to generate variants of a parent binding agent to provide multiple variants from which to select variants with improved developability characteristics.

[0121] To generate a derivative library, the donor DNA of one or more recovered clones is mutated to provide a second repertoire of binders. Similarly, the DNA encoding the parent binder is mutated to generate variants. Mutations may be the addition, substitution, or deletion of one or more nucleotides. Mutations alter the sequence of the encoded binder by the addition, substitution, or deletion of one or more amino acids. Mutations may be focused on one or more regions, such as one or more CDRs of an antibody molecule, to provide a repertoire of binders of a common structural class that differ in one or more regions of diversity, as described elsewhere herein.

[0122] Generally, manipulation and / or modification of nucleic acid sequence can be carried out at DNA or RNA level.Therefore, unless the context requires otherwise, reference to DNA herein can be generalized to include other equivalent nucleic acids (e.g., RNA).Therefore, providing, isolating, or mutating RNA encoding binding agent is an alternative to providing, isolating, or mutating DNA encoding binding agent.RNA is optionally used to generate cDNA.

[0123] Generating a derivative library includes isolating nucleic acid encoding a binding agent from one or more recovered clones (e.g., isolating donor DNA or its encoded RNA), introducing mutations into the nucleic acid (e.g., DNA) to provide a derivative population of donor DNA molecules that encode a second repertoire of binding agents, and introducing the derivative population of donor DNA molecules into cells to create a derivative library of cells containing DNA encoding the second repertoire of binding agents.

[0124] Isolation of nucleic acid (e.g., donor DNA) encoding a binding agent may involve obtaining and / or identifying DNA or RNA from a clone. Such methods may include amplifying the DNA encoding the binding agent from the recovered clone and introducing mutations, for example, by PCR. DNA is sequenced and mutated DNA is synthesized.

[0125] Alternatively, mutations may be introduced into the donor DNA of one or more recovered clones by inducing DNA mutagenesis within the clone. Thus, a derivative library may be created from one or more clones, for example, via endogenous mutation in avian DT40 cells, without the need for DNA isolation. Alternatively, the gene encoding the binder may be present in the genome, and mutagenesis is performed by introducing oligonucleotides with short homology arms. It has been shown that transfection efficiencies of up to 45% can be achieved by using an 80-bp single-stranded oligonucleotide to repair a defective GFP gene (Igoucheva, O., Alexeev, V., Yoon, K., 2001. Targeted gene correction by small single-stranded oligonucleotides in mammalian cells. Gene Ther. 8(5), 391-399). 55 ,Liang,et al(2017).Enhanced CRISPR / Cas9-mediated precise genome editing by improved design and delivery of gRNA,Cas9 nuclease,and donor DNA.J.Biotechnology,241,136-146 56 ).

[0126] Antibody display is particularly useful for generating derivative libraries. Once antibody genes are isolated, various mutagenesis approaches (e.g., error-prone PCR, oligonucleotide-directed mutagenesis, chain shuffling) can be used to create display libraries of related clones from which improved variants can be selected. For example, using chain shuffling of DNA encoding a population of selected VH clones, the oligoclonal mix or population can be subcloned into vectors encoding the preferred antibody format and encoding an appropriately formatted repertoire of VL chains. 57Alternatively, and again using the VH example, an oligomix or population of VH clones can be introduced into a population of eukaryotic cells encoding and expressing a population of appropriately formatted light chain partners (e.g., VL-CL chains for association with IgG- or Fab-formatted heavy chains). The VH population can arise from any of the sources discussed above, including B cells of immunized animals or scFv genes from selected phage populations. In the latter example, cloning selected VHs into a repertoire of light chains can combine chain shuffling and reformatting (e.g., into IgG format) in one step.

[0127] Cell surface display of binders Retention of binders on the cell surface is a hallmark of display libraries because it provides a physical association between the binder and the encoding DNA, facilitating DNA recovery after physical isolation of cells expressing binders with desired properties. Surface display (or simply "display") of binders may also be referred to as binder display or surface expression. The level of binder surface display reflects its expression level and its maintained level of display after prolonged exposure to high concentrations on the cell surface. In the methods described herein, the relative levels of binder surface display are compared among clones expressing different binders and used to identify binders that have a low tendency for self-association, high solubility, and can be formulated in aqueous buffer solutions at concentrations suitable for pharmaceutical use. Therefore, the display level can be used to select clones with desired properties. Thus, surface display of binders on display libraries exhibits unique characteristics that can be used to select for developable properties.

[0128] A variety of means for immobilization on the cell surface can be used. The binding agent may comprise or be linked to a membrane anchor, such as a transmembrane domain, for extracellular presentation of the binding agent. This may involve membrane localization signals, such as GPI recognition sequences, or direct fusion of the binding agent to a transmembrane domain, such as the transmembrane domain of the PDGF receptor. 58 .

[0129] Other methods for achieving cell surface binding agent retention include indirect association of the binding agent with another cell surface-retained molecule expressed within the same cell. This associated molecule may itself be part of a heterodimeric binding agent, such as a tethered antibody heavy chain associated with a light chain partner that is not directly tethered. International Publication No. WO 2015 / 166272 (incorporated herein by reference) describes various techniques for retaining expressed binding agents on their host cells, including methods that enable a combination of secreted expression and membrane display. Thus, some of the expressed binding agent may be retained on the cell surface, while other copies of the same binding agent are secreted from the same cell in soluble form. The cell retains most (e.g., 80% or more, or 90% or more) of the binding agent displayed on the cell surface, with a minority being secreted into the medium. Optionally, the binding agent is only retained on the cell surface and is not secreted in soluble form.

[0130] As illustrated in Example 1, an antibody heavy chain may be fused to a PDGFR TM domain and expressed with its cognate light chain for surface display of a full-length immunoglobulin, e.g., IgG. The gene encoding the intracellular binding agent or its polypeptide subunit (e.g., an antibody heavy chain) may contain DNA encoding a leader sequence for secretion to the cell surface via the endoplasmic reticulum (ER). The gene encoding the binding agent may also contain DNA encoding a membrane anchor, such as a transmembrane domain, e.g., the TM domain of a mammalian (e.g., human) protein. Alternatively, the gene may contain DNA encoding a post-translational membrane anchor attachment signal, such as a glycosylphosphatidylinositol (GPI) anchor. The C-terminal region of a polypeptide binding agent is usually selected for membrane anchor attachment or TM domain fusion.

[0131] Methods for influencing the level of cell surface expression include, for example, using a promoter to control the level at which the binding agent is expressed from its encoding DNA, and methods for doing so are described elsewhere herein. Alternatively, the level of surface expression can be controlled by influencing the degree to which the DNA encoding the expressed binding agent is spliced ​​to the exon encoding the transmembrane domain, as described, for example, in International Publication No. WO2015128509 (Glenmark Pharmaceuticals). This approach is also exemplified herein; see Examples 8a and 8b, which illustrate expression systems that can be usefully used in the methods of the present invention in which various levels of surface expression of the binding agent are desired.

[0132] In an embodiment of the present invention, the binding agent is expressed in a cell, transported to the cell membrane, and retained on the cell surface as a membrane protein. For example, the binding agent may comprise one or more polypeptides having at least one transmembrane domain or membrane anchor. For example, if the polypeptide binding agent comprises an antibody heavy chain (or a portion thereof) and an antibody light chain (or a portion thereof) and comprises a transmembrane domain or membrane anchor linked to the heavy and / or light chain, the binding agent is synthesized in the ER, buds into the Golgi apparatus, and is transported to the cell surface in intracellular vesicles that fuse with the plasma membrane. Thereafter, the binding agent is retained by its transmembrane domain or membrane anchor and presented extracellularly. Thus, incorporation of the binding agent into the membrane occurs intracellularly before transport to the cell surface. Where applicable, assembly of multisubunit binding agents (e.g., antibodies comprising separate heavy and light chains, or portions thereof) can also typically occur intracellularly.

[0133] The level of cell surface display may be measured in copy number (number of displayed binders per cell). Several methods are available, including comparison to calibration beads and Scatchard plots of ligand concentration and receptor occupancy. 59~61With knowledge of the cell radius and certain assumptions regarding the available volume or surface area, this can be used to estimate the concentration or density, respectively (Example 3). At a general level where the present invention is concerned, copy number relates to the concentration achieved, with higher concentrations seen as the display level increases. The relationship between copy number and cell surface concentration is also affected by cell size. Displaying the same number of binders on small and large cells will result in different concentrations, as the antibody will occupy a larger volume in the larger cell (see Example 3, which compares cell size with the concentration achieved).

[0134] Many recombinant expression systems are available, and the absolute number of binding agents displayed on the cell surface will vary from system to system. Those skilled in the art will be able to appropriately calibrate the methods of the present invention depending on the range of display levels observed when using different cells (e.g., different sizes), different promoters, different induction mechanisms, different splicing mechanisms, transport efficiencies, etc. However, the following guidelines are provided by way of example.

[0135] In a weakly active promoter, the binding agent may be displayed on the cell surface at a copy number ranging from 100 to 100,000 per cell. The number of binding agents per cell may be at least 100, at least 1,000, or at least 10,000. The number of binding agents per cell may be up to 1,000, up to 10,000, or up to 100,000. In preferred embodiments, the copy number is no more than about 80,000, no more than about 60,000 per cell, no more than about 50,000 per cell, or no more than about 40,000 per cell. The number of binding agents per cell may be at least 100, at least 1,000, or at least 10,000. Thus, the copy number may be, for example, in the range of 1,000 to 60,000 per cell. This may be about 10,000, about 50,000, or about 60,000.

[0136] In a highly active promoter, the binding agent may be displayed on the cell surface at a copy number ranging from 100,000 to 10,000,000 copies per cell. The number of binding agents per cell may be at least 100,000, or at least 1,000,000 copies. The number of binding agents per cell may be up to 1,000,000 copies, or up to 10,000,000 copies. This may be approximately 1,000,000 copies.

[0137] Of course, even within a single clone, the exact number of binders on different cells will vary, but such variation in copy number between clones is small compared to the variation in copy number between clones that the selection and enrichment methods described herein exploit. Exemplary values ​​and ranges represent approximate average (mean) copy numbers. The copy number may also be the average (mean) copy number of cells in a population. When selecting for affinity, a small copy number is preferred to increase binding stringency and enrich for high-affinity clones, while a large copy number is preferred when selecting for developability, e.g., the solution properties of the binder.

[0138] The copy numbers in a library of cells expressing a binding agent can range from relatively low numbers (e.g., 10,000, 50,000, 100,000, or 250,000 copies per cell) to relatively high numbers (e.g., 1,000,000 copies per cell). The copy numbers quoted here are for guidance only and are based on copy numbers observed in libraries of HEK cells in suspension culture, with a radius of approximately 10 μm. 95 A polypeptide displayed at a given density on the surface of a large cell will be displayed at a higher copy number than if it were displayed at the same density on a small cell, and the absolute copy numbers observed in libraries generated from large or small cells may vary accordingly. See Example 3b.

[0139] When practicing the methods of the invention, one is typically interested in the relative levels of presentation compared between clones, where what is important is not the absolute number of binders (absolute copy number per cell), but the ability to rank or distinguish clones according to the different levels at which binders are displayed on the cell surface.

[0140] The surface display levels of a representative sample of clones from the library can be determined and used to estimate the average (mode, mean, and median) and spread (range) of surface display levels present among the clones in the library.

[0141] Any suitable method can be used to determine the level of polypeptide present on cell surface, and compare their relative amounts.A preferred method for measuring the relative difference in the display of binding agent between cells is to expose cells to a drug that carries a detectable (e.g., fluorescent) label, allow the drug to bind to the binding agent, and detect the relative amount of label on cells, where stronger signals (e.g., more fluorescent units) from the detectable label indicate higher display levels of binding agent on cells.If the detectable label is fluorescent, fluorescence-activated cell sorting (FACS) can be used to perform the selection.Alternatively, or in addition to FACS, magnetic bead selection can be used.

[0142] Generally, agents that bind to the constant region of the binder are selected; all binders can be labeled equally regardless of their sequence. For antibodies and other binders that contain an Fc region, it is convenient to label them with an agent that binds to the Fc region. For example, if the binder is an IgG antibody, the cells can be contacted with a detectable substance that binds to the IgG Fc region, such as a labeled anti-IgG antibody. Where appropriate, the method can be adapted, for example, when a library contains binders that differ in the sequence of their Fc region, and an agent that binds to a non-diverse portion of the Fc or other region of the binder molecule can be used. A peptide expression tag (e.g., hemagglutinin (HA), c-Myc) can be incorporated into the binder polypeptide (e.g., incorporated into the antibody scaffold), allowing the displayed binder to be detected by detecting the agent bound to the tag. This has already been described in the context of selecting correctly assembled antibodies to normalize antigen-binding signals based on antibody expression levels. 24、62 The agent for detecting the surface display level of a binder may be used alone, i.e., without simultaneous detection or selection of other characteristics, such as antigen binding or target specificity. Thus, in some embodiments, detecting the binder display level using an agent that binds to the constant region of the binder does not involve detecting the binder using a labeled target. In other embodiments, multiparameter selection may be performed. A wide range of surface display levels may be exhibited by the binder-expressing clones of the library, reflecting a high level of inter-clone variation. For a representative sample of cells from the library, after detection, e.g., by FACS, the surface display level can be plotted against the frequency at which that display level was observed. Some embodiments of this are described in Example 10, with reference to Figure 31 and exemplified therein. The surface display level or modal surface display level detected for the sample, as well as the spread of display levels and their deviation from the median or mode, may be observed and / or calculated. Modal surface display can be conveniently visualized as the highest peak in a plot of surface display level versus cell number.

[0143] The spread of surface display levels observed in a population of clones may be represented by the degree of dispersion from the modal value (mode). A population of clones may exhibit wide variation in the level of surface display among clones, especially if the binders exhibit large variations in structure (e.g., different primary sequences), resulting in some clones displaying binders at much higher densities than others. While some populations of clones may exhibit lower inter-clonal variation in surface display levels, the method may still be used to set a threshold display level to distinguish higher display clones from lower display clones. One statistical measure of spread is the interquartile range, which is the difference between the upper quartile (third quartile) and the lower quartile (first quartile). The copy number at the third quartile (represented as the lower limit of the upper quartile) may differ from the copy number at the first quartile (represented as the upper limit of the lower quartile) by 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, or at least 5-fold.

[0144] In some embodiments, the range of surface display levels, and the fold difference in surface display between the aforementioned reference points, may be observed to decrease with each round of selection for surface display levels, as the population of cells becomes increasingly enriched for clones with higher surface display of the binder. In some embodiments, the mode may be observed to gradually increase with each enrichment for higher surface display. Following selection, methods for determining or observing an increase in surface display level in a population of binders may include determining or observing an increase in the average (e.g., median or mode) copy number, measured, for example, by FACS or similar methods described herein. A selected population (or clones thereof) may be observed to have an average surface display level that is higher than the average surface display level of an unselected population or clones expressing a comparison polypeptide, e.g., the average surface display level of clones expressing the parent binder from which the population was derived. For example, the mode or median copy number of a selected clone or selected population may be at least 5%, at least 10%, at least 20%, or at least 25% higher than the comparison or parent copy number, and the clone may be selected based on its improved surface display level.

[0145] To select clones expressing higher surface display of the binder, or to select a population of clones enriched for clones encoding binders with higher surface display, cells can be sorted into a collection fraction and a discard fraction according to the level of surface display of the binder on the cells. Cells with surface display above a predetermined threshold are sorted into a collection fraction, and cells with surface display below a predetermined threshold are sorted into a discard fraction. Surface display is optionally the only criterion for selection / enrichment during this step. This can be facilitated by using a detectable agent to label all binders, optionally without detecting target recognition by the binder.

[0146] By discarding the fraction of cells with lower surface display of the binder, clones expressing binders with poor developability properties are depleted from the population. By selecting all or a portion of the remaining cells, a selected cell population is obtained that is enriched for clones expressing binders with higher surface display relative to other clones. In this way, the selected cells, clones, or populations can be identified as having good developability properties compared to the starting population or library and compared to unselected cells, clones, or populations.

[0147] As mentioned above, the absolute number of binders on cell surface varies, and the threshold value suitable for the system being used will be determined by those skilled in the art.For example, the threshold value can be predetermined, for example, based on the first test sample of the library, to select a certain percentage of clones in the population that express the highest surface display of binders.The threshold value can be set, for example, to select the top 50%, top 30%, top 25%, top 20%, top 15%, top 10% or top 5% of cells.

[0148] After determining the surface display level of representative samples of cells from library, and observing or calculating the mode, median, spread and / or interquartile range of copy number of sample, can set appropriate threshold copy number.When using FACS, this will correspond to the threshold fluorescence intensity per cell.For example, if the FACS threshold for cell collection is set to collect library cells with the median fluorescence intensity corresponding to the mode of sample, then cells with this fluorescence level or higher will be collected.To further enrich surface display level, the FACS threshold for cell collection can be set to collect library cells with fluorescence intensity corresponding to the third quartile fluorescence intensity of cells in sample.

[0149] The threshold may represent a number of binders displayed per cell of at least about 100,000, at least about 500,000, or at least about 1,000,000.

[0150] Selection and enrichment Clonal selection, cell selection, or population selection may involve physically separating the clone, cell, or population from other clones or cells, or from a wider population or library. The selected clone, cell, or population may be provided in isolated form.

[0151] When the selection involves physical separation of multiple cells or clones, this will typically involve the generation of a collection fraction and a discard fraction. The collection fraction will be enriched for clones displaying binders with the properties selected in the method. Enrichment refers to an increase in the relative abundance of these clones in the population. Enrichment is relative to the pool of cells or clones, e.g., the library or starting population, prior to the selection step. By discarding a portion of the cells (e.g., cells with lower surface display of binders, cells with lower affinity, etc.) during selection, clones expressing binders with less desirable properties (e.g., poor developability properties) are eliminated. Thus, as a result of the selection step, the relative abundance of cells / binders with less desirable properties is reduced in the population. If desired, all or a portion of the selected cells (collected fraction) may be subjected to further selection methods, and optionally, one or more clones are selected for individual culture alone. Optionally, one or more selected cells or binders or the selected population may be used to create a derivative library, as described elsewhere herein.

[0152] When selecting a pool of clones, the operator of the method may obtain a set percentage or proportion of the "top" or "best" clones. An embodiment of this principle is discussed in detail with reference to selecting for high surface display. It should be understood that the same principle applies when selecting for other properties, such as binding to target or non-target molecules. Thus, when selecting clones expressing binders that recognize a target (e.g., affinity selection), the operator may select a set percentage or proportion of clones with the highest binding to the target (e.g., as measured by the amount of detectable label bound to the clones in a method using a labeled target). When selecting clones expressing binders that exhibit reduced (or absent) non-specific binding to non-target molecules, the operator may select a set percentage or proportion of clones with the lowest binding to non-target molecules (e.g., as measured by the amount of detectable label bound to the clones in a method using a labeled non-target molecule, such as any of the various multireactive probes mentioned herein). Generally, when positively selecting for a "good" property (e.g., surface display level, affinity for target), clones with a high level of that property relative to other clones will be selected, and when selecting against a "bad" property (e.g., non-specific binding), clones with a low level of that property relative to other clones will be selected.

[0153] As discussed elsewhere herein, the selection threshold may be determined by the operator according to the situation at hand or may be guided by an initial evaluation of a sample of the population (e.g., a representative sample of clones from a library). The threshold may be set to enrich the clones with the highest signal levels, e.g., the top 50%, top 30%, top 25%, top 20%, top 15%, top 10%, or top 5% of cells, for the desired properties (e.g., based on the amount of attached detectable label, representing the level of surface display or the level of target binding, or based on interparticle distance determined by AC-SINS, or based on monomer production in solution). The top % of selected clones become the collected fraction, and the other clones are discarded. To select for undesirable properties, for example, based on the amount of detectable label bound to the multireactive probe or the degree of retardation on the matrix for detecting nonspecific binding, a threshold may be set to enrich for clones with the lowest signal levels, for example, to select the percentage of clones with the least amount of detectable label bound or the least degree of retardation on the matrix (e.g., 50%, 30%, 25%, 20%, 15%, 10%, or 5%). The exemplary percentages of collection are, of course, guidelines only, and the exact values ​​are not critical as long as the operator adheres to the principles of the method.

[0154] Quantifying and Measuring Differences Various methods of the invention include comparing the properties of binders and / or their encoding clones, for example, with respect to properties such as display level, binding, concentration (e.g., critical concentration), solubility limit, etc. Methods may include identifying binders that are better relative to other binders in the population (e.g., relative to one or more others, or relative to the population average (mean)) and / or have improved one or more properties relative to the parent molecule from which they are derived. Comparisons may also be made against a benchmark binder, which in the example of an antibody might be the NIST RM 8671 antibody (Saro D et al, Developability Assessment of a proposed NIST monoclonal antibody 37 ). Desirable properties for selection and improvement are discussed elsewhere herein, and include higher solubility, higher critical concentration, lower non-specific binding, and / or higher affinity for the target. Relative terms such as "greater" or "lesser," "higher levels" or "lower levels," "better" or "worse," "more" or "lesser" generally refer to differences observed in a relevant experimental context that allow binders or clones to be distinguished based on their properties. The differences may be statistically significant. The differences may optionally be quantified in terms of percentages, such as a difference of at least 10%, at least 25%, or at least 50%. Alternatively, fold differences may optionally be considered, such as at least 1.5-fold, at least 2-fold, at least 3-fold, at least 5-fold, at least 10-fold, at least 20-fold, or at least 100-fold.

[0155] target binding The binder may be selected for binding to a target molecule of interest, which may optionally be another polypeptide such as a receptor, an enzyme, and / or a disease-associated polypeptide such as a tumor-associated antigen. Other target molecule classes include nucleic acids, carbohydrates, lipids, and small molecules. Exemplary binders and targets are detailed elsewhere herein. A classic example is a library of antibody molecules, which may be screened for binding to a target antigen of interest. Other examples include screening a library of TCRs against a target MHC:peptide complex, or screening a library of MHC:peptide complexes against a target TCR.

[0156] Selection for clones encoding binders (cognate binders) that recognize the target may include contacting a display library described herein with the target, thereby exposing the binder to the target and allowing recognition of the target by the cognate binder (if present), and detecting whether the target is recognized by the cognate binder. One or more clones that display the cognate binder may then be selected.

[0157] The target may be provided in a soluble form. The target may be labeled for easy detection; for example, the label may carry a fluorescent label or may be biotinylated. Cells expressing a target-specific binder may be identified using a directly or indirectly labeled target molecule, in which the binder captures the labeled molecule. For example, cells bound to a fluorescently labeled target via a binder:target interaction can be detected and sorted by flow cytometry to isolate the desired cells. Selection involving cytometry requires a target molecule that is directly fluorescently labeled or labeled with a molecule that can be detected with a secondary reagent; for example, a biotinylated target can be added to cells and binding to the cell surface detected using fluorescently labeled streptavidin, such as streptavidin-phycoerythrin. Another possibility is to immobilize the target molecule or secondary reagent that binds to the target on a solid surface, such as magnetic beads or agarose beads, to allow enrichment of cells that bind to the target. For example, cells that bind to biotinylated targets via binder:target interactions can be isolated on streptavidin-coated substrates, such as streptavidin-coated beads. Magnetic beads are convenient for capturing cells bound to biotinylated antigens by magnetically recovering the beads. The optimal target concentration can be determined in advance or experimentally by using a wide range of concentrations and comparing with background controls.

[0158] The method for selecting target binding includes sorting cells into a collection fraction and a waste fraction according to the level of bound target on the cells, whereby cells with bound target above a predetermined threshold are sorted into the collection fraction, and cells with bound target below a predetermined threshold are sorted into the waste fraction. The threshold may be set relative to a negative control that does not display a cognate binder. In FACS, a negative control peak is typically observed when sorting cells that display cognate and non-cognate binders. The threshold may be set so that all cells that display fluorescence at a statistically significantly higher level than the negative control are sorted into the collection fraction, or a higher threshold may be selected to achieve greater confidence in selecting cognate binders and better enriching clones that display cognate binders. As already discussed with respect to determining cell surface display of binders, calibration may be performed using a cell sample to determine a suitable threshold. Enrichment of binders from non-binders can be achieved. Methods for enriching higher affinity binders are further described elsewhere herein.

[0159] Selection against a target may be incorporated as an additional step before or after other methods of the present invention, or may be included within them. Constructing a library of variants by mutagenesis of the starting domain to improve aspects of developability, such as solubility, low specificity, or optimal FcRn binding, may impair target binding in some library members (e.g., when mutagenesis of contact CDRs is performed). For example, clones selected for binder display level may then be selected against the target. Simultaneous determination of surface display level and target binding is also possible by using simultaneous selection of clones displaying the cognate binder at high surface display levels. Such methods may include the use of an agent incorporating a detectable label to determine binder display level, as described elsewhere herein, and may further include exposing the binder to the target, where the target is labeled with a second agent incorporating a second detectable label to enable detection of target binding. When fluorescent labels are used, FACS may be used to simultaneously sort cells for both binder display and target recognition. Labels may be selected that emit at different wavelengths, allowing these different signals to be distinguished.

[0160] Thus, the methods of the invention may involve simultaneous detection of binder presentation levels and target binding levels, each using a different detectable label. In other embodiments, target binding is detected without determining binder presentation levels on the cells, e.g., detecting the binder using a labeled target does not involve detecting binder presentation levels using an agent that binds to the constant region of the binder.

[0161] After detection of target recognition by the cognate binder, cells of selected clones containing DNA encoding the cognate binder may be recovered. The DNA encoding the binder may then be identified, amplified, and / or provided in isolated form to obtain the DNA encoding the binder that recognizes the target.

[0162] Multiple Selection The idea of ​​performing multiple selections in parallel can be extended to co-sorting cells based on any two or more characteristics described herein. In the above discussion, simultaneous co-selection of cells is performed by simultaneously determining the surface display level of the binder and the level of target binding by the binder, and co-selecting clones that display a higher surface display of the cognate binder. Other methods of the present invention can also be employed in parallel. The methods are described below: (i) Surface presentation level (ii) the level of non-specific binding to non-target molecules (iii) the level of target binding (iv) the level of FcRn binding; and co-selecting clones accordingly. FACS allows parallel selections to be performed using multiple labels that fluoresce at different wavelengths.

[0163] Advantages and synergies can be obtained by performing multiple types of selection in series or in parallel, for example, by combining selection for solubility with selection against nonspecific binding. Each selection applied to a population of clones in a library generates evolutionary pressure favoring variants that meet that selection criterion (e.g., high surface display levels). Repeated selection for a single parameter can drive evolution toward this characteristic (e.g., high solubility) at the expense of other qualities (e.g., affinity for target binding) that are at risk of being depleted or lost. 31 This can occur, for example, when a mutation that increases solubility also decreases affinity. A judicious combination of selection methods can guide the evolution of a population toward clones expressing polypeptides with multiple desired characteristics, allowing the identification of polypeptides that perform optimally (or at least acceptably) across the full range of required requirements for a polypeptide drug.

[0164] Selection for non-specific binding can be carried out before or after selection for increased surface display levels.For example, a method for distinguishing or ranking binders according to their solubility in solution and / or resistance to self-association, and / or a method for enriching for binders that exhibit higher solubility and / or higher resistance to self-association in solution can be carried out to obtain a population of selected clones, and then the selected population can be screened for clones that express binders that exhibit a low tendency to bind to one or more non-target molecules, thereby identifying one or more clones that also have non-specific binding.

[0165] In some embodiments, screening for non-specific binding can be simultaneously performed with screening for surface display level. For example, double exposure of cells to (i) an agent carrying a detectable label (e.g., fluorescently labeled anti-Fc) for binding all displayed binders, and (ii) one or more non-target molecules carrying different detectable labels (e.g., different fluorescent wavelengths) (e.g., heparin or other molecules that prevent non-specific binding) can be performed. Double staining of clones allows clones to be distinguished based on both surface display level and non-specific or off-target binding. A sorting threshold can be set to collect cells that exhibit higher surface display level and less binding to non-target molecules. This eliminates or at least reduces the prevalence of poorly developed antibodies in selected clones.

[0166] Similarly, selection for binding to non-target molecules (eg, heparin or other molecules for which non-specific binding is avoided) can be combined with selection for FcRn binding in the methods described herein.

[0167] In some situations, selecting for one characteristic (e.g., selecting for a high level of surface display) will enrich for binders with multiple beneficial qualities as a result of the combined effects of certain aspects of their exploitability. For example, some clones may express multireactive binders that interact with non-target molecules (e.g., proteoglycans) on the cell surface—for example, they can nonspecifically bind heparin. In higher eukaryotic cell display systems, the binder-expressing cell clones themselves may express the same or similar non-target molecules (e.g., proteoglycans endogenous to the cell), with the resulting effect being that the surface-displayed binder interacts nonspecifically with one or more molecules on the cell on which it is displayed. This can lead to the binder being internalized within the cell and therefore exhibiting a lower level of surface display, resulting in the deselection of the expressing clone. In such cases, selecting clones that display a higher level of surface display of the binder will concurrently enrich or select for clones expressing binders with better exploitability on multiple fronts, e.g., both being more soluble and less subject to nonspecific binding.

[0168] Another example of this is that by enriching for binding agents that have a lower tendency for self-aggregation, the method can facilitate the selection of binding agents that exhibit a lower degree of immunogenicity in vivo by selecting out those binding agents that tend to form immunogenic aggregates when used in pharmaceutical formulations.

[0169] In addition to this, as mentioned above, further dimensions of parallel selection can be incorporated by including further labeled detection agents, such as labeled FcRn, labeled targets, other labeled multireactive probes, etc. The rapid availability of a range of different labels (e.g., fluorophores of different wavelengths) and the ability of FACS machines to perform multiplexed detection and sorting can aid in the design of such parallel selections.

[0170] Affinity selection In selecting binders to a target, it is often useful to be able to select binders on the basis of affinity, allowing enrichment for clones expressing binders with high affinity for target binding.

[0171] Affinity is generally expressed as Kd, the equilibrium dissociation constant. Kd is the ratio k(off) / k(on) for the interaction between a binder and its target (e.g., between an antibody and its antigen). The Kd value is related to the concentration of the binder, and the lower the Kd value (lower concentration), the higher the affinity of the binder. A binder that specifically recognizes its target can be referred to as a cognate binder, in the manner of an antibody that recognizes its antigen. The recognition of a target by a cognate binder is desirably a high-affinity interaction. The Kd:target interaction of a binder can be less than 1 μM, preferably less than 10 nM.

[0172] The methods of the present invention can include enriching a population of cells that encode (and display) binders with higher affinity to a target. The stringency of selection can be enhanced by using eukaryotic cells in which the binder is displayed at a relatively low copy number to drive selection for affinity. A method for selecting binders that bind to a target of interest can employ a library of higher eukaryotic cell clones, each containing DNA encoding a binder, where the binder is displayed on the cell surface, and the encoded binder is expressed from a weakly active promoter and / or expressed on the cell surface at a relatively low copy number. This can be due to expression driven from a weakly active promoter, or as a result of transcript instability or suboptimal splicing, translation, surface transport, or retention on the cell surface.

[0173] Presentation of a dense suspension of higher eukaryotic cells (e.g., 10 7 / ml high level of surface presentation (e.g., 6 × 10 5The binding sites / cells present a relatively high concentration of antibody (10 nM in this example). In that context, even when a low concentration of antigen is used to drive selection stringency, e.g., 0.1 nM, the high concentration of antibody drives association, limiting the relative enrichment between high- and low-affinity clones (Examples 8a and 8b). Input concentrations of binder below 1 nM may be preferred. Even at lower cell densities, there is the potential problem of target rebinding in the presence of high densities of immobilized binder. This problem is particularly well recognized and documented in surface-based affinity assays such as surface plasmon resonance (BIAcore manual). Thus, copy numbers can range, for example, from 100 to 100,000 per cell. In preferred embodiments, the copy number is about 60,000 or less per cell, optionally about 50,000 or less per cell, or about 40,000 or less per cell. To facilitate detection, the copy number can be at least 100, at least 1,000, or at least 10,000 per cell. Thus, the copy number can range, for example, from 1,000 to 60,000 per cell. Copy numbers and methods for determining copy number in the context of surface display of binding agents are discussed elsewhere herein.

[0174] The library is exposed to the target (e.g., by adding the target to a suspension of cells expressing the binding agent of the library), bringing the binding agent into contact with the target and thus allowing recognition of the target by the cognate binder, if present. Cells displaying the cognate binder are bound to the target. By using a limited concentration of the target, binders with higher affinity are preferentially bound by the target. Cells displaying binders that do not recognize the target or that recognize targets with lower affinity will not bind to the target or will display fewer molecules of target per cell compared to cells displaying binders with higher affinity. The target-bound cells are then isolated, thereby allowing selection of a population of cells enriched for cells displaying the cognate binder.

[0175] The selection procedure can optionally be repeated using decreasing concentrations of target to progressively increase the stringency of selection and increase the degree of enrichment for higher affinity clones. Prior to further enrichment for high affinity binders, mutations can be introduced into the binders of the selected population to generate variants. The generation of derivative libraries is described elsewhere herein.

[0176] The concentration of target used may be below the Kd of the interaction of the binding agent sought to be isolated by the method.

[0177] One or more clones having the desired affinity for the target are then selected, and optionally the encoding DNA can then be recovered and the binding agent expressed from the individually cultured recombinant cells as described elsewhere herein.

[0178] To achieve low-level surface display on cells, the binding agent gene can be expressed at low levels, for example, by operably linking to a weakly active promoter, or by transcript instability or suboptimal splicing, translation, transport to the surface, or retention on the cell surface.Inducible promoters and other controllable expression systems are described in detail elsewhere herein.For example, see Example 8a or Example 8b.

[0179] Library A collection of cell clones, each containing recombinant DNA encoding a binding agent, together form a library. The diversity of the library is a function of the number of different binding agents encoded by the clones. In drug discovery, it is advantageous to provide a large and diverse library to maximize the chances of identifying a binding agent that meets all desired criteria. Each clone in the library can be generated by integrating DNA encoding a binding agent into cellular DNA to form a recombinant cell, as described elsewhere herein. The DNA can be introduced into many cells in parallel to generate a population of recombinant cells, each encoding at least one binding agent from a diverse repertoire. After integration of the donor DNA into the cellular DNA, the resulting recombinant cells are cultured to allow their replication, generating clones of cells from each initially produced recombinant cell. Thus, each clone is derived from a single original cell into which the donor DNA was integrated (e.g., an integration site created by a site-specific nuclease or other method described herein). Libraries according to the present invention can contain at least 100, 10 3 , 10 4 , 10 5 , 10 6 , 10 7 , 10 8 , 10 9 or 10 10 The clones may be included.

[0180] A library consistent with the present invention may have one or more of the following characteristics: Diversity. The library has at least 100, 10 3 , 10 4 , 10 5 , 10 6 , 10 7 , 10 8 or 10 9 A number of different binding agents can be encoded and / or expressed. The binding agents of different sequences constitute a repertoire.

[0181] Uniform integration. A library can consist of clones containing donor DNA integrated at a defined locus or at a limited number of defined loci in cellular DNA. Thus, each clone in the library preferably contains donor DNA at a defined locus or at at least one of the defined loci. Preferably, clones contain donor DNA integrated at one or two defined loci in cellular DNA. As described elsewhere herein, the integration site can be a recognition sequence for a site-specific nuclease. Integrating donor DNA to produce recombinant DNA is described in detail elsewhere herein and can produce different results depending on the number of integration sites. If there is a single potential integration site in the cells used to generate the library, the library will be a library of clones containing donor DNA integrated at a single defined locus. Thus, all clones in the library contain the binder gene at the same position in cellular DNA. Alternatively, if there are multiple potential integration sites, the library can be a library of clones containing donor DNA integrated at multiple and / or different defined loci. Preferably, each clone in the library contains donor DNA integrated at a first and / or second predetermined locus. For example, the library can include clones in which donor DNA is integrated at a first predetermined locus, clones in which donor DNA is integrated at a second predetermined locus, and clones in which donor DNA is integrated at both the first and second predetermined loci. In a preferred embodiment, clones in the library have only one or two predetermined loci, but if desired for a particular application, donor DNA can be integrated at multiple loci. Thus, in some libraries, each clone can contain donor DNA integrated at any one or more of several predetermined loci, for example, 3, 4, 5, or 6 predetermined loci.

[0182] For libraries containing binder subunits integrated at distinct sites, clones of the library contain DNA encoding a first binder subunit integrated at a first defined locus and DNA encoding a second binder subunit integrated at a second defined locus, and the clones express multimeric binders comprising the first and second subunits.

[0183] Uniform transcription. The relative levels of transcription of binders among different clones in a library are kept within controlled limits by donor DNA integration at a controlled number of loci and at the same locus (defined locus) in different clones. Relatively uniform transcription of binder genes leads to comparable levels of binder expression on or from clones in the library. Binders displayed on the surface of cells in the library can be identical (have the same amino acid sequence) to other binders displayed on the same cell. A library can consist of clones of cells each displaying a single member of a binder repertoire, or clones displaying multiple members of a binder repertoire per cell. Alternatively, a library can contain some clones displaying a single member of a binder repertoire and some clones displaying multiple members (e.g., two) of a binder repertoire. Preferably, clones in a library express one or two members of a binder repertoire.

[0184] For example, a library of eukaryotic clones according to the invention may contain at least 10 3 , 10 4 , 10 5 , 10 6 , 10 7 , 10 8 or 10 9The cells can express a repertoire of different binding agents, e.g., IgG, Fab, scFv, or scFv-Fc antibody fragments. Each cell contains donor DNA integrated into the cellular DNA. The donor DNA encodes the binding agent and can further contain genetic elements for selection of cells into which the donor DNA has been integrated. Cells of the library can contain DNA encoding an exogenous site-specific nuclease.

[0185] A binding agent displayed on the surface of a cell in a library can be identical (have the same amino acid sequence) to other binding agents displayed on the same cell. A library can consist of clones of cells that each display a single member of a repertoire of binding agents, or clones that display multiple members of a repertoire of binding agents per cell. Alternatively, a library can contain some clones that display a single member of a repertoire of binding agents and some clones that display multiple members (e.g., two) of a repertoire of binding agents. Thus, a library according to the invention can contain clones encoding two or more members of a repertoire of binding agents, where the donor DNA is integrated at overlapping defined loci or multiple independent defined loci.

[0186] If the corresponding clone expresses only one binding agent, it is easiest to identify the corresponding coding DNA for the binding agent. Typically, a molecule of donor DNA will encode a single binding agent. A binding agent may be a multimer, and the molecule of donor DNA will contain multiple genes or open reading frames corresponding to the various subunits of the multimeric binding agent.

[0187] As mentioned, the library according to the present invention contains at least 100, 10 3 , 10 4 , 10 5 or 10 6 , 10 7 , 10 8 , 10 9 or 10 10Each of the binding agents can encode a number of different binding agents. If the binding agent is a multimer, diversity can be provided by one or more subunits of the binding agent. A multimeric binding agent can combine one or more variable subunits with one or more constant subunits, where the constant subunits are the same (or have more limited diversity) across all clones in the library. When generating a library of multimeric binding agents, combinatorial diversity is possible, where a first repertoire of binding agent subunits can be paired with any of a second repertoire of binding agent subunits.

[0188] These and other features of the libraries according to the present invention are further described elsewhere herein, and examples of suitable libraries and methods for their construction and use are also set forth in WO2015 / 166272 (Iontas Limited), the contents of which are incorporated herein by reference. Suitable loci can be identified for targeted integration of encoding DNA into cell chromosomes, and several examples are known in the art. The AAVS locus can be used as exemplified herein. Other suitable integration sites include the ROSA26, HPRT, and FUT8 loci (e.g., in CHO cells). While targeted integration can have advantages, random integration is a suitable alternative and can be used in many situations. Aside from nuclease-mediated targeted integration, other methods for generating libraries of surface-expressing polypeptides include transfecting eukaryotic cells with vectors encoding the polypeptides, e.g., using lentiviruses, adenoviruses, adeno-associated viruses, or transposons, or using genomically embedded recombinase sites, such as endogenous cryptic recombinase sites, such as Flp, Bxb2, or phi recombinase sites. Libraries created by these or any other techniques can be used in the methods described herein. The present invention extends to libraries either in pure form as a population of library clones in the absence of other eukaryotic cells, or mixed with other eukaryotic cells. The other cells can be the same type of eukaryotic cell (e.g., the same cell line) or different cells. Additional advantages can be obtained by combining two or more libraries according to the present invention, or by combining a library according to the present invention with a second library or a second population of cells, to facilitate or expand screening, or for other uses described herein or that will be apparent to those skilled in the art.

[0189] Host cells into which DNA encoding a library according to the invention, one or more clones obtained from the library, or a binding agent from the library has been introduced can be provided in cell culture medium, the cells can be cultured and then concentrated to form a cell pellet for convenient transport or storage.

[0190] The library will typically be provided in vitro. The library may be in a container such as a cell culture flask containing the library cells suspended in culture medium, or in a container containing a pellet or concentrated suspension of eukaryotic cells containing the library. The library may comprise at least 75%, 80%, 85%, or 90% of the eukaryotic cells in the container. Selection steps may be performed on the library in mixed culture, thus facilitating high throughput.

[0191] Instead of co-culturing a mixture of library clones, it may be convenient to culture the clones individually, each in its own separate flask or other container. Individual cultures may be used when relatively small numbers of clones are to be compared, such as when a parent binder has been mutated to generate one or more variants (e.g., up to 10 variants) and the invention is used to compare the quality of the variants relative to the parent and / or each other.

[0192] The selection method described herein can be applied to naive libraries, i.e., libraries that have not undergone affinity-based selection.Therefore, the method can be used to enrich the library for clones with high solubility and / or lower non-specific binding (or conversely, to deplete the library for clones that exhibit low solubility and / or higher non-specific binding) before carrying out affinity-based selection.Then, such a library that has been "pre-selected" for more exploitable clones is highly suitable for carrying out affinity-based selection using the target of interest.Compared to the clones selected from the library that have not been pre-screened for exploitability, the clones obtained in the affinity selection step are more likely to exhibit good solubility, high critical concentration, low non-specific binding, and / or other exploitable qualities.

[0193] eukaryotic cell The eukaryotic cells according to the invention are preferably 12 x 10 6 A higher eukaryotic cell is defined herein as a cell having a genome larger than that of Saccharomyces cerevisiae, which has a genome size of 100 base pairs (bp). A higher eukaryotic cell is, for example, a cell having a genome size of 2 x 10 7 The eukaryotic cells of the present invention may have a genome size greater than 100 base pairs. This includes, for example, mammalian, avian, insect, or plant cells. The eukaryotic cells of the present invention preferably lack a cell wall. Preferably, they are not yeast or other fungal cells. Preferably, the cells are mammalian cells, such as mouse or human. The cells may be primary cells or cell lines. Chinese hamster ovary (CHO) cells are commonly used for antibody and protein expression, although any alternative stable cell line may be used in the present invention. HEK293 cells are used in some examples herein.

[0194] Display of binders using higher eukaryotes, such as mammalian cells, has advantages because antibody production for research, diagnostic, and therapeutic applications is typically performed in these cells. Performing drug discovery in mammalian cells, which exhibit the same expression environment and post-translational modifications, would provide a better indication of potential problems or benefits for downstream manufacturing, allowing early identification of clones with optimal expression characteristics. In contrast, bacterial and yeast cells do not fully recapitulate the glycosylation, expression, and secretion machinery of higher eukaryotes. Therefore, display in mammalian cells may help identify clones with better display levels or stability characteristics, which could have implications for future research uses or downstream manufacturing. The ability to display large libraries of antibodies on the surface of mammalian cells would allow for the direct screening of millions of clones for binding and display properties with the potential for using manufacturing cell lines during the discovery phase of antibody development.

[0195] CHO cell lines were first isolated in 1957. 63 Derivatives of this cell line are the producer cell lines for the majority of therapeutic antibodies. 64 For example, Herceptin (an anti-HER-2 antibody approved to treat breast cancer) is produced at over one metric ton per year by expression in CHO cells. CHO cells have the advantage for producing products for human administration because, compared to human cells, they do not propagate most human pathogenic viruses. In addition, they allow the incorporation of foreign DNA into their genome and grow rapidly and robustly. The properties of antibodies and other polypeptide binders, including biophysical properties, stability, pharmacokinetics, and immunogenicity, depend on glycosylation acquired within the cell's secretory pathway, such as the endoplasmic reticulum (ER) and Golgi apparatus. 65The glycosylation profile of antibodies can be influenced by post-translational modifications such as mannose (Man), galactose (Gal), fucose (Fuc), N-acetylglucosamine (GlcNAc), N-acetylgalactosamine (GalNAc), and sialic acid. Here, monosaccharide units such as mannose (Man), galactose (Gal), fucose (Fuc), N-acetylglucosamine (GlcNAc), and sialic acid are covalently attached to specific amino acids. "O-linked" or "N-linked" glycosylation refers to either glycans attached to the oxygen atom of serine or threonine residues or to the amide nitrogen atom of asparagine residues. Glycosylation complexity can be introduced by either linear or branched glycosylation and the atomic position and conformation of the glycosidic bond (e.g., α or β) at various positions within the monosaccharide unit. The exact nature of an antibody glycosylation profile can be influenced by the host cell line used for expression. 65 Therefore, during antibody and therapeutic protein screening, it is advantageous to produce recombinant proteins in a host cell line that is as close as possible to the final production cell line. This ensures that the post-translational modifications of the screened polypeptides, and therefore their properties, are identical or as similar as possible to those that will be obtained during their large-scale production. Since the majority of human therapeutic antibodies are produced in CHO cells, it is therefore advantageous to perform higher eukaryotic display in a CHO host cell line. Example 12 demonstrates the developability-based selection of candidate polypeptide drugs in a CHO cell library.

[0196] In the method and use of the present invention, generally, the plurality of cell clones can be a library of at least 1000 clones, optionally cultured together in the same culture medium in a single container.It can be a naive library, i.e., encoding a repertoire of binders that have not been previously selected for binding to a target.This will often be the case in early-stage discovery, but it may be desirable to use a library that encodes a repertoire of binders that is the result of one or more previous rounds of selection for binding to a target.For example, the selection output of a phage display library can be introduced into a eukaryotic cell library.

[0197] The invention is described herein with particular reference to mammalian cells, and mammalian cells are used to illustrate the invention in the Examples. However, it should be understood that other higher eukaryotic cells may be used instead, unless the context requires otherwise. For example, insect or chicken cells may be used.

[0198] Binder A "binding agent" according to the present invention is a binding molecule that represents a specific binding partner of another molecule. Typical examples of specific binding partners are antibody-antigen and receptor-ligand. Many principles of the present invention extend to polypeptides that cannot be classically considered "binding agents," such as enzymes, cofactors, clotting factors and their inhibitors, and complement factors and their inhibitors. Often, such polypeptides represent candidate clinical drugs that are desirable to produce on a large scale and / or provide at high concentrations. Therefore, their developability is an important consideration. For example, factor VIII is used in hemophilia, but has developability issues that must be addressed. The developability aspects of the present invention can be applied to all such polypeptides. Therefore, unless the context otherwise requires, the present invention should not be construed as limited to classical specific binding molecules such as antibodies, but should be understood to generally extend to polypeptides designed to interact with one or more other molecules.

[0199] The present invention relates to binding agents that are polypeptides, i.e., polymers of amino acids, that are expressed in a cell from encoding DNA and, optionally, undergo post-translational modifications such as cleavage, glycosylation, etc. Binding agents can comprise short peptides, e.g., on the order of 10-30 amino acids. Binding agents can also be longer polypeptides, optionally comprising multiple subunits.

[0200] The binding agent preferably comprises a polypeptide of mammalian, e.g., human, origin. The binding agent can comprise a human antibody (optionally a chimeric antigen receptor (CAR) comprising a human antibody), a human TCR or other receptor, or other human polypeptide. The binding agent can be a soluble peptide or polypeptide (e.g., a cytokine, chemokine, complement protein or complement regulator, enzyme (including enzymes for industrial use), or blood clotting factor, e.g., factor VIII). The binding agent can be a mammalian (e.g., human) membrane protein or a soluble mammalian (e.g., human) protein / peptide engineered to contain one or more TM domains or other membrane anchors. The binding agent can be a naturally occurring polypeptide or (frequently) a synthetic variant. For example, a library of human factor VIII polypeptides can include binding agents having at least 70% amino acid sequence identity (e.g., at least 80%, or at least 90%) with human factor VIII.

[0201] The repertoire of binding agents encoded by the library will typically share a common structure and have one or more regions of diversity. Thus, the library allows for the selection of members of a desired structural class of molecules, such as peptide or scFv antibody molecules. Thus, in a library or population of binding agents in accordance with the invention, the binding agent polypeptides can share a common structure (e.g., related secondary and / or tertiary structure, optionally including regions of highly similar or identical amino acid sequences—"constant regions") and have one or more regions of amino acid sequence diversity—"variable regions."

[0202] This can be illustrated by considering a repertoire of antibodies. These can be antibodies of a common structural class, e.g., IgG, Fab, scFv-Fc, or scFv, that differ in one or more regions of their sequence. Antibodies typically have sequence variability in the complementarity-determining regions (CDRs), the regions primarily responsible for antigen recognition. A repertoire of binding agents in the present invention can be a repertoire of antibody molecules that differ in one or more CDRs, for example, all six CDRs, or one or more specific CDRs, such as heavy chain CDR3 and / or light chain CDR3.

[0203] Antibodies and other binding agents are described in more detail elsewhere herein. However, the capabilities of the present invention extend beyond antibody display to include the display of libraries of peptides or engineered proteins, including receptors, ligands, individual protein domains, and alternative protein scaffolds. 66~68 Examples are polypeptides with monomeric binding domains, such as DARPins and lipocalins, affibodies, and adhirons. The present invention can also be used with complex multimeric binders. For example, T cell receptors (TCRs) are expressed on T cells and have evolved to recognize peptides presented in complex with MHC molecules on antigen-presenting cells. Libraries encoding and expressing repertoires of TCRs can be generated and screened to identify binding to MHC-peptide complexes.

[0204] For multimeric binders, donor DNA encoding the binder can be provided as one or more DNA molecules.For example, if the VH and VL domains of individual antibodies are expressed separately, they can be encoded on separate molecules of donor DNA.Donor DNA is integrated into cellular DNA at multiple integration sites, i.e., the VH binding gene at one locus and the VL binding gene at a second locus.The method of introducing donor DNA encoding separate binder subunits is described in more detail elsewhere herein and in WO2015 / 166272 (Iontas Limited), which is incorporated herein by reference.Alternatively, both subunits or portions of multimeric binders can be encoded on the same molecule of donor DNA, which is integrated into a specific locus.

[0205] The binding agent can be an antibody or a non-antibody protein that contains an antigen-binding site. The antigen-binding site can be provided to confer binding to a desired target by means of the placement of a peptide loop on a non-antibody protein scaffold, such as fibronectin or cytochrome B, or by randomizing or mutating amino acid residues in a loop within the protein scaffold. (Haan & Maggos (2004) BioCentury, 12(5):A1-A6 69、70 Protein scaffolds for antibody mimetics are disclosed in WO / 0034784, in which the inventors describe proteins (antibody mimetics) containing fibronectin type III domains with at least one randomized loop. A suitable scaffold for grafting one or more peptide loops, e.g., a set of antibody VH CDR loops, can be provided by any domain member of the immunoglobulin gene superfamily. The scaffold can be a human or non-human protein.

[0206] The use of antigen-binding sites in non-antibody protein scaffolds has been reviewed previously (Wess, L. In: BioCentury, The Bernstein Report on BioBusiness, 12(42), A1-A7, 2004). Typical are proteins with a stable backbone and one or more variable loops, where a loop or loops of amino acid sequence are specifically or randomly mutated to create an antigen-binding site that exhibits binding to the target antigen. Such proteins include the IgG-binding domain of protein A from S. aureus, transferrin, tetranectin, fibronectin (e.g., the 10th fibronectin type III domain), and lipocalins. Other approaches include, for example, "knottins" and small constrained peptides based on cyclotide scaffolds. 71 Given their small size and complexity, particularly in relation to the correct formation of disulfide bonds, the use of eukaryotic cells for the selection of novel binders based on these scaffolds may be advantageous. Given the general functionality of these peptides, libraries of binders based on these scaffolds may be advantageous for generating small, high-affinity binders with particular application in blocking ion channels and proteases. WO 2017 / 118761 (Iontas Limited) described a library of binding members, each comprising a fusion protein containing a donor diversity scaffold domain, such as a cysteine-rich protein, inserted into a recipient diversity scaffold domain, such as an antibody constant or variable domain. Such binders and libraries as described in WO 2017 / 118761 may be used in the present invention, and the document is incorporated herein by reference. Thus, in some embodiments, a binder according to the present invention is a "knotbody" comprising a cysteine-rich protein inserted into an antibody variable domain. See Example 15 herein.

[0207] In addition to the antibody sequence and / or antigen-binding site, the binding agent can contain other amino acids, for example, to form a peptide or polypeptide, such as a folding domain, or to confer another functional feature to the molecule in addition to its ability to bind to an antigen. The binding agent can carry a detectable label or be conjugated to a toxin, targeting moiety, or enzyme (e.g., via a peptidyl bond or linker). For example, the binding agent can contain a catalytic site (e.g., in an enzyme domain) as well as an antigen-binding site, which binds to the antigen and thus targets the catalytic site to the antigen. The catalytic site can inhibit the biological function of the antigen, for example, by cleavage.

[0208] Optionally, the binders in the library are a population of polypeptides whose thermal stability (e.g., as determined by melting temperature) does not predict the binder's solubility or resistance to self-association in solution. This lack of correlation between thermal stability and solubility in solution / resistance to self-association can be applied when considering all binders in the library that have a solubility or critical concentration of at least 10 mg / ml (methods for determining either are described elsewhere herein). It is possible to determine the critical concentration of at least 10 mg / ml. 3 , at least 10 per cell 4 , at least 10 per cell 5 , or at least 10 per cell 6 It can be applied when considering all binders with surface display (any of the methods for determining this described elsewhere herein, e.g., FACS gating). Optionally, it can be applied when considering the total population of surface-expressed binders in a library.

[0209] antibody Antibodies are preferred binding agents. They can be whole antibodies or immunoglobulins (Ig), which have four polypeptide chains—two identical heavy chains and two identical light chains. The heavy and light chains form a pair, each having a VH-VL domain pair that contains an antigen-binding site. The heavy and light chains also contain a constant region, a light chain CL, and heavy chain CH1, CH2, CH3, and sometimes CH4. The two heavy chains are connected by disulfide bridges at a flexible hinge region. An antibody molecule can contain a VH and / or a VL domain.

[0210] The most common naturally occurring form of antibody molecules is IgG, a heterotetramer consisting of two identical heavy chains and two identical light chains. The heavy and light chains are composed of modular domains with conserved secondary structure consisting of a four-stranded antiparallel beta sheet and a three-stranded antiparallel beta sheet stabilized by a single disulfide bond. Each antibody heavy chain has an N-terminal variable domain (VH) and three relatively conserved "constant" immunoglobulin domains (CH1, CH2, and CH3), while the light chain has one N-terminal variable domain (VL) and one constant domain (CL). Disulfide bonds stabilize the individual domains and form covalent bonds linking the four chains in a stable complex. The VL and CL of the light chain associate with the VH and CH1 of the heavy chain, and these elements can be expressed alone to form Fab fragments. The CH2 and CH3 domains (also called "Fc domains") associate with another CH2:CH3 pair to form a tetrameric Y-shaped molecule with the variable domains from the heavy and light chains at the ends of the "Y." The CH2 and CH3 domains are involved in interactions with effector cells and complement components within the immune system. Recombinant antibodies have previously been expressed in the IgG format or as Fabs (consisting of a VH:CH1 dimer and a light chain). Additionally, artificial constructs called single-chain Fvs (scFvs) can be used, consisting of DNA encoding the VH and VL fragments genetically fused with DNA encoding a flexible linker.

[0211] IgG is one of the preferred classes of antibodies for therapeutic use. The advantage of using higher eukaryotic cells, especially mammalian cells, is that antibodies can be studied in IgG format. This allows drug discovery and screening to be carried out directly in the production cell type used for IgG production.

[0212] The binding agent may be a human antibody molecule, and therefore, where constant domains are present, these are preferably human constant domains.

[0213] The binding agent can be in the form of a smaller antibody molecule, such as an antibody fragment or a single-chain antibody molecule. For example, the antibody molecule can be an scFv molecule consisting of a VH domain and a VL domain connected by a linker peptide. In the scFv molecule, the VH and VL domains form a VH-VL pair, and the complementarity determining regions of the VH and VL together form an antigen binding site.

[0214] Other antibody fragments that contain an antibody antigen-binding site include: (i) a Fab fragment consisting of the VL, VH, CL, and CH1 domains; (ii) a Fd fragment consisting of the VH and CH1 domains; (iii) a Fv fragment consisting of the VL and VH domains of a single antibody; and (iv) a dAb fragment consisting of the VH or VL domain. 72~74 (v) isolated CDR regions; (vi) an F(ab')2 fragment, in which the bivalent fragment comprises two linked Fab fragments; and (vii) an scFv, in which the VH and VL domains are linked by a peptide linker that allows the two domains to associate and form an antigen-binding site. 75、76 (viii) bispecific single-chain Fv dimers (PCT / US92 / 09965) and (ix) gene fusions (WO94 / 13804, 77 These include, but are not limited to, bispecific antibodies, which are multivalent or multispecific fragments constructed by the synthesis of Fv, scFv, or bispecific antibody molecules that may be stabilized by incorporating disulfide bridges linking the VH and VL domains. 78 .

[0215] A variety of other antibody molecules have been engineered that contain one or more antibody antigen-binding sites, including, for example, Fab2, Fab3, bispecific antibodies, triabodies, tetrabodies, and minibodies (small immune proteins). Antibody molecules and methods for their construction and use have been described. 79 .

[0216] Other examples of binding fragments are Fab', which differs from Fab fragments only by the addition of a few residues at the carboxyl terminus of the heavy chain CH1 domain including one or more cysteines from the antibody hinge region, and Fab'-SH, which is a Fab' in which the cysteine ​​residue(s) of the constant domains bear a free thiol group.

[0217] A dAb (domain antibody) is a small, monomeric antigen-binding fragment of an antibody, i.e., the variable region of an antibody's heavy or light chain. VH dAbs occur naturally in camelids (e.g., camels, llamas) and can be produced by immunizing camelids with a target antigen, isolating antigen-specific B cells, and directly cloning dAb genes from individual B cells. dAbs can also be produced in cell culture. Their small size, good solubility, and temperature stability make them particularly physiologically useful and suitable for selection and affinity maturation. Camelid VH dAbs are being developed for therapeutic use under the name "nanobodies™."

[0218] Synthetic antibody molecules can be created by expression from genes generated by means of oligonucleotides synthesized and assembled in a suitable expression vector, for example, as described by Knappik et al. or Krebs et al. 80、81 .

[0219] Bispecific or bifunctional antibodies form a second-generation monoclonal antibody in which two different variable regions are combined in the same molecule 63Their use has been demonstrated in both the diagnostic and therapeutic fields due to their ability to recruit new effector functions or to target several molecules on the surface of tumor cells. When bispecific antibodies are used, they can be produced in a variety of ways, for example, traditional bispecific antibodies prepared chemically or from hybrid hybridomas, or they can be any of the bispecific antibody fragments described above. 82 These antibodies can be synthesized by chemical methods. 83、84 or somatic cell methods 85、86 ), but also preferentially obtained by genetic engineering techniques that allow heterodimerization to be forced, thus facilitating the process of purification of the desired antibodies. 87 Examples of bispecific antibodies include BiTE™ technology, in which the binding domains of two antibodies with different specificities can be used directly via a short, flexible peptide, combining two antibodies on a short, single polypeptide chain. Bispecific antibodies and scFvs can be constructed without an Fc region, using only variable domains and potentially reducing the effects of anti-idiotypic reaction.

[0220] In some embodiments of the present invention, the binding agents are bispecific antibodies, whose encoding clones contain two different antibody heavy chains and, optionally, two different antibody light chains, or preferably, a common light chain. Successful heterodimeric pairing between heavy chains results in cell surface display of the bispecific antibody, with each antibody containing a heterodimeric pair of heavy chains, optionally paired with a light chain, and optionally with a common light chain (i.e., the light chains paired with each heavy chain have the same amino acid sequence). When an Fc region is included, the present invention can be used to evaluate Fc sequence variants that can improve heterodimerization. In previous studies in which the Fc region was engineered to improve heterodimerization, the development profile was impaired by the alterations. 67 The present invention provides the opportunity to screen such variants for developability (eg, solubility and multispecificity) as well as their heterodimerization potential.

[0221] Bispecific antibodies can be constructed as whole IgG, as bispecific Fab'2, as Fab'PEG, as diabody, or else as bispecific scFv. Furthermore, two bispecific antibodies can be linked using routine methods known in the art to form tetravalent antibodies.

[0222] In contrast to bispecific whole antibodies, bispecific diabodies may also be particularly useful. Bispecific antibodies (and many other polypeptides, such as antibody fragments) of appropriate binding specificity can be easily selected. If one arm of the bispecific antibody is held constant, e.g., with specificity directed against an antigen of interest, a library can be created in which the other arm is varied to select antibodies of appropriate specificity. Bispecific whole antibodies can be generated by alternative engineering methods, such as those described in Ridgeway et al. (Protein Eng., 9, 616-621, (1996)).

[0223] An alternative form of bispecific antibody is a mAb, which comprises an immunoglobulin in which the CH3 loop region has been engineered to provide an antigen-binding site. 2 ("mAb squared") molecules (see, for example, WO2006072620, WO2008003103, WO2008003116). The modified CH3 region is referred to as Fcab. One binding specificity is provided by the antibody antigen-binding site of the Fv region, and a different binding specificity (or additional valency) is provided by the binding site in the Fcab.

[0224] The libraries of the present invention can be used to select antibodies that bind to one or more antigens of interest. Selection from libraries is described in detail elsewhere herein. After selection, the antibodies can be engineered into different formats and / or engineered to include additional functions. For example, the selected antibodies can be converted into different formats, such as one of the antibody formats described above. The selected antibodies and antibodies comprising the VH and / or VL CDRs of the selected antibody molecules are embodiments of the present invention. The antibodies and their encoding nucleic acids can be provided in isolated form.

[0225] Antibody fragments can be obtained starting from antibody molecules by methods such as digestion with enzymes, for example, pepsin or papain, and / or by cleavage of disulfide bridges by chemical reduction. Alternatively, antibody fragments can be obtained by techniques of genetic recombination well known to those skilled in the art, or else by peptide synthesis, for example by means of an automated peptide synthesizer, or by nucleic acid synthesis and expression.

[0226] It is possible to take monoclonal and other antibodies and use recombinant DNA technology techniques to generate other antibodies or chimeric molecules that bind to target antigens. Such techniques can involve introducing nucleic acid (e.g., DNA) encoding the immunoglobulin variable region, or CDRs, of an antibody to the constant regions, or constant regions and framework regions, of a different immunoglobulin. See, for example, EP-A-184187, GB 2188638A, or EP-A-239400, and numerous subsequent references.

[0227] Antibody molecules may be selected from the library and then modified, for example, the in vivo half-life of the antibody molecule may be increased by chemical modification, such as PEGylation, or by incorporation into liposomes.

[0228] The binding agents can optionally comprise antibody variable domains exhibiting sequence diversity in one or more complementarity determining regions. The binding agents can also, or alternatively, comprise antibody constant regions or Fc regions, optionally exhibiting sequence diversity. The function of the Fc region is further described below.

[0229] FC area The binder may be a polypeptide comprising an Fc region. The binder may be an antibody, knotbody, or other polypeptide, optionally a fusion protein, comprising an Fc region. The Fc region may be or may include the constant region of the binder, i.e., have a very similar or identical amino acid sequence when compared between binders in the library. In some cases, the constant domain of an antibody (e.g., the Fc region, or the CL or CH1 domain) may be or may include a variable amino acid sequence, thus exhibiting sequence diversity across the repertoire of binders in the library. The sequence diversity may optionally be in the CH3 domain of the Fc. For example, the binder may be an Fcab or mAb, where the binding loops of the Fcab are diverse in the library. 2 It may contain mAb. 2 The binding agents can comprise either non-variant or diverse antigen-binding sites of diverse Fcab and Fv regions. The amino acid sequence diversity of the binding agents can be restricted to the Fc region and optionally to the CH3 domain.

[0230] Using the display libraries of the invention, libraries of Fc domains can be screened, optionally simultaneously, for altered function and developability criteria.

[0231] Various engineering approaches have been taken to manipulate the interaction of the Fc domain with its interaction partners. For example, the "knobs-into-hole" approach modifies two paired Fc sequences such that their co-expression from cells (or their expression in co-cultured cells) primarily results in heterodimer formation between the two mutant Fcs domains, which is advantageous in the generation of bispecific antibodies. Unfortunately, such mutations can have implications for developability. 88 The present invention can be used to evaluate Fc variants, including Fc domains containing candidate "knobs-into-holes" mutations, for potential developability.

[0232] Fc engineering has also been used to alter affinity or specificity with Fc gamma receptors, for example, to create "null variants" with reduced Fc gamma receptor interaction. Mutation of antibody constant domains and selection of variants with desired binding qualities can reduce stability and manufacturability. Furthermore, the present invention can be used to evaluate libraries of variant Fc domains and enrich for those with improved properties.

[0233] Modifications have also been made to increase or decrease the interaction of Fc with FcRn in order to positively or negatively modify half-life. 89 For example, the triplet of mutations M252Y / S254T / T256E (the so-called "YTE mutation") has an increased IgG half-life, prolonging half-life and reducing administration frequency and cost.

[0234] It is understood that introducing mutations into the Fc domain can lead to aggregation and poor developability, and also have detrimental effects on the biophysical properties of the mutants. For example, Borrok et al. (2017) 90review mutations that affect interaction with FcRn and with other Fc receptors. They describe antibodies that combine mutations that increase half-life (M252Y, S254T, T256E, so-called "YTE mutations") and antibodies that reduce interaction of the CH2 domain with Fc gamma receptors (L234F, L235E, P331S, so-called TM mutations). 90 We describe an antibody combining the TM-YTE and TM-YTE complexes. Compared to the wild-type, this TM-YTE variant exhibited reduced thermal stability, greater conformational flexibility, increased self-association, poorer solubility, and a poorer aggregation profile. By selecting candidate mutations, we were able to create a new FQQ-YTE variant (L234F / L235Q / K322Q / M252Y / S254T / T256E) with significantly improved conformational and colloidal stability while retaining an extended half-life and lack of antibody-dependent cell-mediated cytotoxicity and complement-dependent cytotoxicity activity.

[0235] The present invention provides the opportunity to simultaneously screen multiple variants for altered binding properties of the Fc domain, along with exploitability criteria such as self-aggregation or reduced specificity. Thus, in various embodiments of the present invention, a binder can comprise an Fc region exhibiting sequence diversity, for example, in one or more amino acid residues in the CH3 domain. Binders exhibiting sequence diversity in one or more variable regions outside the Fc domain (e.g., antibody heavy and / or light chain variable domains), optionally with a constant Fc domain of invariant sequence, can be screened. Screening can be performed to identify the effect of the variable region sequence on FcRn interaction (see Example 9). Such effects can be indirect, for example, affecting the conformation of the molecule in the absence of direct binding of the variable region to the receptor, or otherwise affecting more distant binding.

[0236] Screening for FcRn-binding properties of polypeptides can be incorporated into drug discovery. The method can include selecting for optimal pH-dependent FcRn interaction within a library of sequence variants. Described herein is a method for identifying or selecting binding agents with extended or reduced in vivo half-life due to the nature of their interaction with FcRn.

[0237] The method according to the present invention comprises: providing a plurality of eukaryotic (e.g., mammalian) cell clones each containing DNA encoding a binding agent comprising an Fc domain; Culturing the clones in vitro under conditions for presenting the binding agent on the cell surface; exposing the clone to the FcRn receptor at low pH (e.g., about pH 6.0) or neutral pH (e.g., about pH 7.4) to allow recognition of FcRn by the Fc domain; and selecting one or more clones expressing a binder that elutes at a higher pH (e.g., about pH 7.4) and exhibits higher affinity binding at about pH 6.0 compared to binding at a higher pH (e.g., about pH 7.4).

[0238] Selected clones can be obtained from the eluted fractions. Binders of the eluted clones can be identified as having an extended half-life in vivo.

[0239] Alternatively, binders retained after switching to higher pH can be collected if retained binding at higher pH is desired, e.g., due to reduced half-life or for use in a "sweeping antibody" approach. 32 In such cases, the method is providing a plurality of eukaryotic (e.g., mammalian) cell clones each containing DNA encoding a binding agent comprising an Fc domain; Culturing the clones in vitro under conditions for presenting the binding agent on the cell surface; exposing the clone to the FcRn receptor at a low pH (e.g., about pH 6.0) that allows recognition of FcRn by the Fc domain; and washing at a higher pH (e.g., about pH 7.4) and selecting one or more clones expressing binders that exhibit lower or similar affinity binding at about pH 6.0 compared to binding at a higher pH (e.g., about pH 7.4).

[0240] Selected clones can be obtained from the retained fraction that are not eluted by the higher pH wash. Such clones can be eluted at more extreme pHs, above pH 7.4 or below pH 6. Higher affinity binding at lower pH (e.g., about pH 6.0) can be selected for by decreasing the concentration of FcRn used at that pH during selection, thereby increasing the stringency of selection.

[0241] Binders that exhibit a larger difference in affinity between the two pHs can be preferentially selected because they can exhibit a longer maximum half-life. Conversely, if a shorter half-life is desired, clones expressing binders with a smaller (or no significant) difference in affinity for FcRn between the two pHs will be selected. Thus, the method can be adapted to select for either a shorter or longer half-life. Thus, the population of clones is enriched for clones expressing binders with the desired half-life.

[0242] Binding to biotinylated FcRn can be performed at pH 6.0 to confirm that binding has occurred, and eluted samples can be collected after washing with buffers of increasing pH. Alternatively, selection for retention or loss of fluorescent label can be performed using flow sorting. FcRn-based affinity chromatography methods combined with pH gradient elution have been described to characterize antibodies. 91 Such methods can be used with libraries of antibody variants displayed on cells, where clones with desired pH-dependent binding properties can be collected and the antibody genes recovered.

[0243] Some embodiments of the present invention use anti-Fc detection agents to determine the level of surface display of a binding agent. Such agents can still be used in conjunction with binding agents that have diversity in Fc sequences, as long as the diversity does not affect the binding of the detectable agent, e.g., ensuring that the detection agent binds to a region of the Fc with which the binding agents share a common sequence. In alternative embodiments, the binding agent comprises an Fc region that does not exhibit sequence diversity.

[0244] Cell culture and binder expression To provide a repertoire of binders for screening and / or developmental characterization against a target of interest, the library can be cultured to express the binders from the encoding DNA. As discussed, the binders can include a transmembrane domain, a membrane anchor, or can associate with a membrane-binding partner molecule for extracellular display. Culturing cells for binder expression will generally involve incubating in an appropriate culture medium, optionally in suspension culture, and at a temperature conducive to cell growth (e.g., 37°C for mammalian cells). Expression of the polypeptide binder from the encoding DNA is initiated under the control of a promoter (and optionally other elements such as an enhancer), and surface display of the binder begins to be observed after a period of time, and should be detectable within, for example, 12 hours, although a longer period (e.g., 24 or 48 hours) may be required for binder display to reach a final or equilibrium concentration on the cell surface.

[0245] promoter In the cells according to the present invention, the DNA encoding the binding agent is operably linked to a promoter for expression. A heterologous promoter may be used, meaning that it is not the promoter naturally associated with the encoding DNA; for example, the DNA encoding an antibody may be operably linked to a promoter other than that from an immunoglobulin. While it is generally convenient to incorporate a promoter into the cellular DNA, optionally in cis (in the same donor DNA as the sequence encoding the binding agent), an alternative is to express the incorporated binding agent DNA from a promoter endogenous to the host cell.

[0246] High levels of binder presentation can be achieved when the expression of the DNA encoding the binder is under the control of a strong promoter, e.g., a constitutive or inducible promoter in which expression is maximally induced. Conversely, low levels of binder presentation can be achieved when the expression of the DNA encoding the binder is under the control of a weak promoter, e.g., a weak or inducible promoter in which expression is minimally induced or exhibits only basal activity. The strength of the promoter can be quantified using a reporter gene, and the level of reporter expression, e.g., GFP expression, can be detected as quantifiable fluorescence. A weak or weak promoter can exhibit, for example, approximately 1-10% of the activity of a fully active or constitutive promoter, compared to, for example, the CMV promoter.

[0247] A convenient way to control cell surface display of a binding agent is to provide an inducible promoter operably linked to the binding agent-encoding DNA in the cell. Tetracycline-inducible promoters are preferred, and a variety of these are available. Gossen et al. described the first generation rtTA protein (EP 0804565 and Gossen et al., 1995). 92) The VP16 activation domain was fused to a mutant Tet repressor from Escherichia coli to generate the transcriptional transactivator "rtTA," which requires a specific tetracycline (Tc) derivative for specific DNA binding. Doxycycline is an inducer in this system, and its addition to cultured cells in which gene expression is under the control of rtTA can result in a 1000-fold increase in expression from the inducible promoter. A second-generation "TetO / CMV" promoter, designated pTight or Ptet-14, was designed with an optimized 7-TetO spacing and a truncated minimal CMV promoter that exhibits reduced basal expression levels (Clontech: pTRE-Tight Vectors. Clontechniques 2003, 18(3):13-14). The promoter sequence is as follows: TTCGTCTTCACACGAGTTTACTCCCTATCAGTGATAGAGAACGTATGTCGAGTTTACTCCCTATCAGTGATAGAGAACGATGTCGAGTTTACTCCCTATCAGTGATAGAGAACGTATGTCGAGTTTACTCCCTATCAGTGATAGAGAACGTATGTCGAGTTTAC TCCCTATCAGTGATAGAGAACGTATGTCGAGTTTATCCCTATCAGTGATAGAGAACGTATGTCGAGTTTACTCCCTATCAGTGATAGAGAACGTATGTCGAGGTAGGCGTGTACGGTGGGAGGCCTATATAAGCAGAGCTCGTTTAGTGAACCGTCAGATCGCC

[0248] This inducible promoter, or various other regulatable promoters, can be used in the present invention to control binder presentation levels. Any inducible system can be employed if the DNA-binding domain is fused to a protein domain capable of binding to an inducer molecule, resulting in a protein conformational change or a change in affinity for the DNA recognition sequence. This will result in either derepression or activation of transcription, leading to protein expression. For example, in the case of the T-Rex or cumate switch system, binding of the inducer to a repressor protein results in loss of DNA binding and derepression of transcription. Alternatively, binding of the inducer to a DNA-binding domain fused to a transcription activation domain fusion protein results in the Tet-on 92、114 or GAL4 GeneSwitch system 121 As in the case of , this would result in DNA binding and recruitment of transcription factors.

[0249] The tetracycline-inducible promoter system described above was developed by reducing the number of TetO repeats from 1 to 6. 118 This system may be convertible to more tunable, uniform, and titratable induction by rTA or by modification of the rTA protein. Example 8b describes a third-generation inducible tet promoter system that achieves an improved range of expression. Details are illustrated in Figure 37. An alternative inducible expression system is the coumate switch. 119 , T-Rex 120 or GAL4 system 121 may also be adopted.

[0250] Recovery of binding agents and encoding nucleic acids After selecting a desired binding agent or clone from the library, the next general step is to isolate, identify, and / or amplify the nucleic acid (e.g., DNA or RNA) encoding the binding agent. Optionally, it may be desirable to modify the nucleic acid encoding the binding agent, for example, to reconstruct the binding agent and / or to insert the coding sequence into a different vector. The nucleic acid (DNA or RNA) encoding the displayed binding agent can be recovered from the selected cells and cloned into an expression vector, and the encoded polypeptide can be expressed either in a secreted form or for display. This can be done for individual clones.

[0251] When the collection of donor DNA molecules used to create a library contains multiple copies of the same sequence, two or more clones can be obtained that contain the DNA encoding the same binding agent.For example, as described elsewhere herein, when there are two or more recognition sequences for site-specific nucleases, clones can contain the donor DNA encoding two or more different binding agents.Therefore, the diversity of library in terms of the number of different binding agents encoded or expressed can differ from the number of clones obtained.

[0252] The clones in the library preferably contain donor DNA encoding one or two members of the binder repertoire and / or preferably express only one or two members of the binder repertoire. A limited number of different binders per cell is advantageous for identifying clones and / or DNA encoding specific identified binders when screening the library for a given target. This is most simple when the clones encode a single member of the binder repertoire. However, if the clones selected from the library encode a small number of different binders, for example, if the clones can encode two members of the binder repertoire, it is also simple to identify the coding DNA associated with the desired binder. As discussed elsewhere herein, clones encoding one or two binders are particularly convenient to generate by selecting recognition sequences for site-specific nucleases that occur once per chromosome copy in the diploid genome, because diploid cells contain duplicated anchoring loci, one on each chromosome copy, and donor DNA can be integrated at one or both anchoring loci. Thus, each clone in the library can express only one or two members of the repertoire of binders.

[0253] When the binding agent is an antibody molecule, the method can include isolating DNA encoding the antibody molecule from the clonal cells, amplifying the DNA encoding at least one antibody variable region, preferably both the VH and VL domains, and inserting the DNA into a vector to provide a vector encoding the antibody molecule. Multimeric antibody molecules with constant domains can be converted into single-chain antibody molecules for expression in a soluble or secreted format. Antibodies can be displayed in a variety of formats, and regardless of the format in which the antibody is selected, once the antibody gene is isolated, it can be reconstituted in several different formats. Once the VH or VL domain is isolated, it can be recloned into an expression vector containing the necessary partner domains.

[0254] DNA encoding the selected binding agent can be integrated into a host cell chromosome for expression. Expression of recombinant proteins typically involves introducing a gene encoding the desired protein into a cell under the control of a promoter expressed in that cell. For example, the gene can be under the control of the cytomegalovirus enhancer / promoter and introduced into mammalian cells such as commonly used human embryonic kidney 293 (HEK293) cells or Chinese hamster ovary (CHO) cells. Standard methods for introducing expression constructs into host cells are well known. For the production of secreted, soluble proteins, the encoded gene is preceded by a leader sequence that directs the encoded protein to the endoplasmic reticulum. In the absence of a transmembrane domain, the encoded gene is secreted into the culture medium, where it can be purified and concentrated.

[0255] According to any method of the present invention, the desired binding agent can be provided in isolated form in solution, for example, after secretion from host cells stably transfected for expression of the binding agent. The properties of the soluble binding agent can then be tested to confirm its performance, for example, to evaluate developmental properties such as solubility, self-association, nonspecific binding, FcRn interaction, etc., or to evaluate affinity. Suitable assays for determining each of these characteristics are provided in the relevant sections of this document and can be performed to confirm that the binding agent exhibits the desired properties and / or shows improvement in a related characteristic, for example, it is improved compared to the parent molecule.

[0256] Pharmaceutical Formulation of Binder The binding agents obtained using the methods of the present invention may be provided in purified and / or isolated form or may be formulated into compositions containing one or more additional components. The compositions may include suitable carriers, excipients, and other agents incorporated into the formulation to provide improved mobility, delivery, tolerance, etc. Exemplary formulations are described in Remington's Pharmaceutical Sciences. Binders intended for in vivo use may be formulated for the desired route of administration to a patient, for example, in an injectable liquid (optionally, an aqueous solution). Various delivery systems are known and can be used to administer pharmaceutical compositions containing the binding agents. Administration methods include intradermal, intramuscular, intraperitoneal, intravenous, subcutaneous, intranasal, epidural, and oral routes. The compositions may be administered by any convenient route, such as by infusion or bolus injection, absorption through epithelial or mucocutaneous linings (e.g., oral mucosa, rectal and intestinal mucosa, etc.), and may be administered together with other biologically active agents. Administration may be systemic or local.

[0257] The binding agent, or a composition comprising the same, can be contained in a medical container, such as a medicine vial, syringe, intravenous device, or injection device. In embodiments, a kit is provided that includes the binding agent, packaging, and instructions for use in a therapeutic method. The method can include subcutaneous administration to a patient.

[0258] The binder may be formulated into a composition, optionally an aqueous buffer solution, containing the binder at a concentration of at least (or more than) 50 mg / ml, 60 mg / ml, 70 mg / ml, 80 mg / ml, 90 mg / ml, or 100 mg / ml. The binder may be at a concentration of 50-200 mg / ml, e.g., 50-150 mg / ml, e.g., 50-100 mg / ml.

[0259] Many methods for concentrating expressed recombinant proteins are known to those skilled in the art, including column chromatography (e.g., affinity chromatography, ion exchange chromatography) and ultrafiltration. 27 may include:

[0260] term The following numbered paragraphs represent the statement of the invention and are part of the description.

[0261] 1. Using the level of surface display of binders on cultured higher eukaryotic cells in vitro as a predictor of binder solubility in solution and / or their resistance to self-association.

[0262] 2. The use according to item 1, wherein the cultured cells are clones of a display library that express a diverse repertoire of binding agents.

[0263] 3. The use of a culture library of higher eukaryotic cells to select binding agents in vitro for greater solubility in solution and / or a lower tendency to self-associate, The use wherein the library is a library of higher eukaryotic cell clones each containing DNA encoding a binding agent, the encoded binding agents being displayed on the cell surface.

[0264] 4. A method for binder discovery in which the level of surface display of binders on the surface of cultured higher eukaryotic cell clones of a display library is used as a predictor of binders in solution and / or their resistance to self-association.

[0265] 5. A method for differentiating and ranking binding agents according to their solubility in solution and / or resistance to self-association, and enriching for binding agents that exhibit higher solubility in solution and / or higher resistance to self-association, comprising: (i) providing a library of higher eukaryotic clones each containing DNA encoding a binding agent; (ii) culturing the clones in vitro under conditions for expression of the binding agent, wherein the binding agent is displayed on the cell surface; (iii) determining the level of surface display of the binding agent on the clone, optionally by labeling the binding agent with an agent incorporating a detectable (e.g., fluorescent) label; (iv) selecting one or more clones that exhibit higher surface display of the binder compared to other clones; (v) identifying binding agents encoded by one or more selected clones as having good solubility in solution and / or resistance to self-association, and optionally providing the selected clones for use in one or more further screening steps.

[0266] 6. The use according to any one of items 1 to 3, or the method according to item 4 or 5, wherein the binding agent is a transmembrane domain-containing polypeptide.

[0267] 7. The method of clause 4 or clause 5, comprising determining the level of surface display of the binding agent on the clone by labeling the binding agent with an agent having a detectable (e.g., fluorescent) label incorporated therein, wherein the agent binds to a constant region of the binding agent, and optionally, the binding agent comprises an Fc region and the agent binds to the Fc region.

[0268] 8. The method of any of paragraphs 4 to 7, wherein the cells are sorted into a collect fraction and a discard fraction according to the level of surface display of the binder on the cells, whereby cells that exhibit surface display above a predetermined threshold are sorted into the collect fraction and cells that exhibit surface display below the predetermined threshold are sorted into the discard fraction.

[0269] 9. The method of paragraph 8, wherein the waste fraction contains cells expressing a comparison polypeptide having a critical concentration of at least 10 mg / ml, and the collection fraction contains cells expressing a binding agent having a critical concentration at least 1.5 times higher than the comparison polypeptide in the waste fraction.

[0270] 10. The method of paragraph 8 or paragraph 9, wherein sorting is performed by fluorescence activated cell sorter (FACS).

[0271] 11. The method of any of paragraphs 4 to 10, wherein step (ii) comprises culturing the clones of the library as a mixture in a single vessel.

[0272] 12. The method according to any one of items 4 to 10, wherein step (ii) comprises culturing each clone of the library in a separate container.

[0273] 13. The method of any one of items 4 to 12, wherein the binding agent is a sequence variant of the parent binding agent.

[0274] 14. The method of paragraph 13, wherein the parent binding agent is identified as needing improved solubility in solution or resistance to self-association.

[0275] 15. A method comprising generating sequence variants of a parent binding agent and incorporating DNA encoding the sequence variants into cellular DNA of higher eukaryotic cells to provide a library of cellular clones comprising DNA encoding the binding agent; Optionally, the method of paragraph 13 or paragraph 14, wherein the method comprises analyzing a parent polypeptide sequence, identifying one or more amino acid residues predicted to promote self-association and / or reduce solubility, and generating mutations in the one or more amino acid residues.

[0276] 16. The parent binding agent has a critical concentration of less than 50 mg / ml in phosphate buffered saline solution (PBS) and / or a solubility limit of less than 50 mg / ml in phosphate buffered saline solution (PBS); and / or the method of any of items 13 to 15, wherein the method comprises identifying the binding agents encoded by one or more selected clones as having a critical concentration and / or solubility limit at least 1.5 times higher than that of the parent binding agent.

[0277] 17. The method of any of paragraphs 4 to 16, comprising predicting the hydrophilicity of binders based on their surface display level on cell clones and / or identifying binders of one or more selected clones as being more hydrophilic.

[0278] 18. An in vitro method for screening a library of higher eukaryotic cells that exhibit binding agents to enrich the library for cells that express binding agents that have a low propensity to bind to one or more non-target molecules in a mammal in vivo, comprising: (i) providing a library of higher eukaryotic clones each containing DNA encoding a binding agent; (ii) culturing the clones in vitro under conditions for expression of the binding agent, wherein the binding agent is displayed on the cell surface; (iii) exposing the binding agent to one or more non-target molecules to allow binding of the binding agent to the one or more non-target molecules; (iv) discarding cells that have greater binding to one or more non-target molecules; (v) selecting cells that have low binding to one or more non-target molecules to provide a population of selected cells enriched for clones that express binders that have a low propensity to bind to non-target molecules; and optionally and providing the selected population for use in one or more further screening steps.

[0279] 19. (iii) exposing the binding agent to a matrix containing one or more non-target molecules to allow binding to the matrix; (iv) discarding cells that are more strongly attached to the matrix; (v) selecting cells that bind poorly to the matrix to provide a population of selected cells enriched for clones that express binders that have a low propensity to bind non-target molecules; and optionally and providing the selected population for use in one or more further screening steps.

[0280] 20. The method of paragraph 18 or paragraph 19, wherein the non-target molecule comprises one or more components of DNA, heparin, heparan sulfate, chondroitin sulfate, champerone protein, hyaluronic acid, or glycocalyx.

[0281] 21. The method according to any one of items 18 to 20, wherein the binding to the non-target molecule is low-affinity non-specific binding.

[0282] 22. The method of any one of items 18 to 21, comprising culturing the clones of the library as a mixture in a single container and exposing the mixture to a matrix.

[0283] 23. The method according to any one of items 18 to 21, comprising culturing each clone of the library in a separate container.

[0284] 24. The method of any one of items 18 to 23, wherein the binding agent is a sequence variant of the parent binding agent.

[0285] 25. The method of paragraph 24, wherein the parent binding agent is identified as requiring reduced binding to one or more non-target molecules.

[0286] 26. A method comprising generating sequence variants of a parent binding agent and integrating DNA encoding the sequence variants into cellular DNA of higher eukaryotic cells to provide a library of cellular clones comprising DNA encoding the binding agent; Optionally, the method of paragraph 24 or paragraph 25, wherein the method comprises analyzing a parent polypeptide sequence, identifying one or more amino acid residues predicted to promote non-specific binding, and generating mutations in the one or more amino acid residues.

[0287] 27. The method of any of paragraphs 24 to 26, wherein the parent binding agent exhibits significant binding to one or more non-target molecules.

[0288] 28. (iii) A method according to any one of paragraphs 18 to 27, comprising exposing the binding agent to cells or beads presenting one or more non-target molecules.

[0289] 29. The method of paragraph 28, comprising detecting interactions between binding agent-expressing cells and cells or beads presenting one or more non-target molecules.

[0290] 30. The method of claim 29, comprising using AC-SINS to detect interparticle distances and selecting binders presented by cells exhibiting larger interparticle distances.

[0291] 31. The method of paragraph 18, wherein one or more non-target molecules are detectably labeled.

[0292] 32. The method of paragraph 31, wherein one or more non-target molecules are fluorescently labeled.

[0293] 33. The method of paragraph 32, wherein cells are flow sorted by FACS into a collection fraction and a waste fraction according to their level of binding to one or more non-target molecules, whereby cells exhibiting fluorescence from labeled non-target molecules above a predetermined threshold are sorted into a collection fraction and cells exhibiting fluorescence from labeled non-target molecules below a predetermined threshold are sorted into a waste fraction.

[0294] 34. The method according to any one of items 4 to 33, wherein expression of the DNA encoding the binding agent is under the control of a strong promoter.

[0295] 35. The method of paragraph 34, wherein the promoter is a constitutive promoter.

[0296] 36. The method of paragraph 35, wherein the promoter is a CMV promoter.

[0297] 37. The method of paragraph 34, wherein the promoter is an inducible promoter that is maximally induced for expression.

[0298] 38. The method according to any one of items 18 to 37, further comprising subsequently carrying out the method according to any one of items 5 to 17.

[0299] 39. The method according to any one of items 18 to 37, further comprising first carrying out the method according to any one of items 5 to 17.

[0300] 40. A method for selecting one or more binding agents to a target, comprising: It includes carrying out the method defined in any of paragraphs 5 to 39, exposing the binder to the target to allow recognition of the target by the cognate binder, thereby causing cells that display the cognate binder to become bound to the target; selecting one or more clones that exhibit a homogenous binder.

[0301] 41. (i) simultaneously determining the level of surface display of the binder and the level of target binding by the binder, and simultaneously selecting clones that exhibit higher surface display of the cognate binder compared to other clones; or (ii) simultaneously determining the surface display level of the binder and the level of non-specific binding to non-target molecules, and simultaneously selecting clones that exhibit binders with higher surface display and lower non-specific binding compared to other clones; or (iii) simultaneously determining the level of target binding and the level of non-specific binding of the binder to non-target molecules, and simultaneously selecting clones that exhibit a cognate binder with less non-specific binding compared to other clones.

[0302] 42. A method for identifying a binding agent that recognizes a target with a desired affinity, comprising: (a) providing an in vitro library of higher eukaryotic cell clones each containing DNA encoding a binding agent, wherein the binding agent is displayed on the cell surface, and the encoded binding agent is expressed from a weakly active promoter and / or is expressed on the cell surface at a copy number ranging from 100 to 60,000 copies per cell; (b) exposing the library to the target to allow recognition of the target by the cognate binder, thereby rendering cells that display the cognate binder bound to the target; (c) isolating the target-bound cells to provide a population of selected cells enriched for cells that display a cognate binder; and optionally, (d) exposing the selected population of cells to one or more rounds of selection on a target, optionally decreasing the concentration of the target to increase the stringency of the selection; and optionally (e) selecting one or more clones that exhibit a cognate binder with the desired affinity for the target.

[0303] 43. Providing a population of selected cells enriched for cells that exhibit a homotypic binder; and then providing binding agent-encoding DNA from a population of selected cells under the control of a highly active promoter in an in vitro library of higher eukaryotic cell clones; Item 43. The method according to Item 42, comprising carrying out the method according to any one of Items 5 to 17.

[0304] 44. Carrying out the method defined in any one of paragraphs 5 to 17 to obtain selected clones with higher surface display of the binding agent, and then expressing the binding agents from a weakly active promoter in an in vitro library of higher eukaryotic clones each containing DNA encoding the binding agent; and 43. The method of claim 42, comprising carrying out the method of claim 42.

[0305] 45. A method for identifying a binding agent that recognizes a target, comprising: (i) providing a library of higher eukaryotic cell clones each containing DNA encoding a binding agent, wherein expression of the binding agent is under the control of an inducible promoter for display on the cell surface; (ii) culturing the cells of the library under conditions in which the activity of the inducible promoter is weak; (iii) exposing the library to the target and allowing recognition of the target by the cognate binder, thereby rendering the cells exhibiting the cognate binder bound to the target; (iv) selecting cells that exhibit a cognate binder to provide a population of selected cells; (v) culturing the selected population of cells under conditions for increased expression of the binding agent from an inducible promoter; (vi) determining the surface display level of the binding agent on the plurality of clones, optionally by labeling the binding agent with an agent incorporating a detectable (e.g., fluorescent) label; (vii) selecting one or more clones that exhibit higher surface display of the binder compared to other clones.

[0306] 46. ​​A method for identifying a binding agent that recognizes a target, comprising: (i) providing a library of higher eukaryotic cell clones each containing DNA encoding a binding agent, wherein expression of the binding agent is under the control of an inducible promoter for display on the cell surface; (ii) culturing the library under conditions for strong expression of the binders from an inducible promoter; (iii) determining the surface display level of the binding agent on the plurality of clones, optionally by labeling the binding agent with an agent incorporating a detectable (e.g., fluorescent) label; (iv) selecting a population of clones that exhibit higher surface display of binders compared to other clones; (v) culturing the selected population under conditions for weak expression of the binding agent from an inducible promoter; (vii) exposing the library to the target and allowing recognition of the target by the cognate binder, thereby rendering the cells displaying the cognate binder bound to the target; (iv) selecting one or more clones that exhibit a cognate binder.

[0307] 47. The method of any one of items 42 to 46, wherein the promoter is a tetracycline-inducible promoter.

[0308] 48. The method of any of paragraphs 42 to 47, wherein the target is labeled with a detectable agent, such as a fluorescent label.

[0309] 49. The method of paragraph 48, wherein the method comprises sorting cells into a collection fraction and a waste fraction according to the level of bound target on the cells, whereby cells exhibiting bound target above a predetermined threshold are sorted into the collection fraction and cells exhibiting bound target below the predetermined threshold are sorted into the waste fraction.

[0310] 50. The method of paragraph 49, wherein sorting is performed by fluorescence activated cell sorter (FACS).

[0311] 51. Determining the sequence of DNA encoding the binding agent from one or more selected clones; 51. The method of any of items 5 to 50, comprising providing an isolated nucleic acid encoding the binding agent.

[0312] 52. Determining the sequence of DNA encoding the binding agent from one or more selected clones; 52. The method of any of paragraphs 5 to 51, further comprising expressing DNA encoding the binding agent in a host cell in vitro under conditions for secretion of the binding agent in soluble form.

[0313] 53. The method of paragraph 52, wherein the secreted binding agent is obtained in a yield of at least 1 mg / ml.

[0314] 54. The method of clause 52 or clause 53, further comprising purifying and / or concentrating the binding agent to obtain an aqueous solution of the binding agent at a concentration of at least 10 mg / ml.

[0315] 55. The method of claim 54, wherein the concentration is at least 50 mg / ml.

[0316] 56. The method of claim 55, wherein the concentration is at least 100 mg / ml.

[0317] 57. The method of any of paragraphs 52 to 56, comprising formulating the binder in a composition comprising a pharmaceutically acceptable excipient.

[0318] 58. The method of claim 57, comprising providing the composition in a pre-filled syringe for injection, optionally in a kit containing one or more additional components such as a product information leaflet containing instructions for administering the composition by needle and / or injection.

[0319] 59. A method for identifying a binding agent that interacts with FcRn, comprising: providing a plurality of higher eukaryotic cell clones each containing DNA encoding a different binding agent having an Fc domain; Culturing the clones in vitro under conditions for presenting the binding agent on the cell surface; exposing the clone to the FcRn receptor at about pH 6.0 and about pH 7.4 to allow recognition of FcRn by the Fc domain; selecting one or more clones that express a binder that exhibits higher affinity binding at about pH 6.0 compared to about pH 7.4, a binder that exhibits lower affinity binding at about pH 6.0 compared to about pH 7.4, or a binder that exhibits about the same affinity binding at about pH 6.0 compared to about pH 7.4; and optionally providing the selected clones for use in one or more further screening steps.

[0320] 60. The method of paragraph 59, comprising selecting one or more clones expressing a binding agent that exhibits higher affinity binding at about pH 6.0 compared to about pH 7.4, and identifying the binding agent encoded by the one or more selected clones as having an increased half-life in vivo.

[0321] 61. The method of clause 59 or clause 60, wherein the binding agents comprise variable domains exhibiting sequence diversity, optionally in one or more complementarity determining regions.

[0322] 62. The method of any of clauses 59-61, wherein the binding agents comprise Fc regions that exhibit sequence diversity, optionally in their CH3 domains.

[0323] 63. The method of any of clauses 59 to 61, wherein the Fc region of the binding agent does not exhibit sequence diversity.

[0324] 64. A clone, a binding agent expressed by the clone, or a nucleic acid encoding a binding agent substantially as described herein and / or identified or selected by a method according to any of paragraphs 1 to 63.

[0325] 65. Higher eukaryotic cell clones each containing DNA encoding a binding agent, the encoding DNA optionally being at a defined locus in the cellular DNA, and the encoded binding agent being expressed on the cell surface at a copy number in the range of 100-1000 copies per cell.

[0326] 66. Use of a library as defined in paragraph 65 for affinity-based selection of binders for a target.

[0327] 67. An in vitro display library of higher eukaryotic cell clones containing DNA encoding a repertoire of binding agents, wherein expression of the binding agents is under the control of a tetracycline-inducible promoter for display on the cell surface.

[0328] 68. A method for producing a library of higher eukaryotic clones containing DNA encoding a repertoire of binding agents, comprising: providing a donor DNA molecule encoding a binding agent, and a higher eukaryotic cell; introducing donor DNA into a cell to create a recombinant cell in which the donor DNA has been integrated into the cellular DNA; creating a vector in which expression of the binding agent is under the control of a tetracycline-inducible promoter for presentation at the cell surface; Culturing the recombinant cells to produce clones; thereby providing a library of higher eukaryotic cell clones containing donor DNA encoding a repertoire of binding agents.

[0329] 69. The method of paragraph 68, wherein the recombinant cell is created by introducing donor DNA into the cell and providing a site-specific nuclease within the cell, wherein the nuclease cleaves a recognition sequence in the cellular DNA, creating an integration site at which the donor DNA is integrated into the cellular DNA, and wherein integration occurs through DNA repair mechanisms within the cell.

[0330] 70. The method of clause 68 or clause 69, further comprising inducing expression of the donor DNA from a tetracycline-inducible promoter and culturing the cells under conditions for expression of the binding agent to obtain display of the binding agent on the cell surface.

[0331] 71. The method of clause 69 or clause 70, further comprising using the library in a method or use as defined in any of clauses 1 to 70.

[0332] 72. The use, method, or library according to any one of paragraphs 1 to 71, wherein the higher eukaryotic cells are mammalian cells.

[0333] 73. The use, method, or library according to any one of paragraphs 1 to 72, wherein the mammalian cells are a human cell line or a CHO cell line.

[0334] 74. The use, method, or library according to any one of paragraphs 1 to 73, wherein the higher eukaryotic cells are in suspension culture.

[0335] 75. The use, method, or library according to any of paragraphs 1 to 74, wherein the DNA encoding the binding agent is integrated at a defined locus in the cellular DNA.

[0336] 76. The use, method, or library according to any one of paragraphs 1 to 75, wherein the binding agent is an antibody.

[0337] 77. The use, method, or library of paragraph 76, wherein the antibody is a full-length immunoglobulin.

[0338] 78. The use, method, or library of paragraph 77, wherein the antibody is an IgG.

[0339] 79. The use, method, or library of any of paragraphs 76 to 78, wherein the antibody comprises a heavy chain and a light chain fused to a transmembrane domain.

[0340] 80. The binding agent is a fusion protein comprising a donor diversity scaffold domain inserted into a recipient diversity scaffold domain, and optionally a partner domain associated with the fusion protein; 80. The use, method, or library of any of paragraphs 1 to 79, wherein the donor diversity scaffold domain comprises a donor scaffold and a donor interaction sequence, and the recipient diversity scaffold domain comprises a recipient scaffold and a recipient interaction sequence.

[0341] 81. The use, method, or library of paragraph 80, wherein the fusion protein is a knotbody comprising a cysteine-rich peptide inserted into an antibody variable domain.

[0342] 82. The use, method, or library of paragraph 81, wherein the knotbody comprises an antibody heavy chain and an antibody light chain fused to a transmembrane domain.

[0343] 83. The use, method, or library of any of paragraphs 1 to 82, wherein the binding agents comprise antibody variable domains exhibiting sequence diversity, optionally in one or more complementarity determining regions.

[0344] 84. The use, method, or library of any of paragraphs 1 to 83, wherein the binding agents comprise Fc regions that exhibit sequence diversity, optionally in their CH3 domains.

[0345] 85. The use, method, or library of any of paragraphs 1 to 84, wherein the binding agent is multispecific, comprising a first binding site for a first target and a second binding site for a second target.

[0346] 86. The library has at least 10 3 86. The use, method, or library according to any one of items 1 to 85, comprising clones.

[0347] 87. The use, method, or library according to any one of paragraphs 1 to 86, wherein the library is a naive library.

[0348] 88. The use, method, or library according to any of paragraphs 1 to 86, wherein the clones of the library have been preselected for binding to a selected target.

[0349] 89. The use, method, or library of paragraph 88, wherein the target is a human polypeptide.

[0350] 90. The use, method, or library of clause 88 or clause 89, wherein clones of the library have been pre-selected for bispecific binding to two different targets.

[0351] Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein which equivalents are intended to be encompassed by the claims appended hereto.

[0352] Other aspects and embodiments of the present invention provide those aspects and embodiments described above with the term "comprising" replaced with the term "consisting of" and those aspects and embodiments described above with the term "comprising" replaced with the term "consisting essentially of."

[0353] It should be understood that the present application discloses all combinations of the above aspects and any of the embodiments described above with each other unless the context requires otherwise. Likewise, the present application discloses all combinations of preferred and / or optional features alone or together with any of the other aspects unless the context requires otherwise.

[0354] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention belongs. Although any compositions and methods similar or equivalent to those described herein can be used in the practice or testing of the method of this disclosure, exemplary compositions and methods are described herein. Any of the aspects and embodiments of the disclosure described herein can also be combined. For example, the subject matter of the dependent or independent claims disclosed herein can be combined in multiples (for example, one or more statements from each dependent claim can be combined into a single claim based on the independent claim on which they depend).

[0355] As used in the specification and claims, the singular forms "a," "and," and "the" include plural references unless the context clearly dictates otherwise. Thus, for example, reference to a "peptide chain" is a reference to one or more peptide chains and includes equivalents thereof known to those skilled in the art.

[0356] All documents and sequence database entries referred to in this specification are incorporated herein by reference in their entirety for all purposes.

[0357] As used herein, "and / or" should be considered as a specific disclosure of each of the two specified features or components with or without the other. For example, "A and / or B" should be considered as a specific disclosure of each of (i) A, (ii) B, and (iii) A and B, as if each were individually set forth herein. [Brief explanation of the drawings]

[0358] Embodiments of the present invention will now be described in more detail with reference to the following drawings:

[0359] [Figure 1] pINT17-BSD, a dual-promoter antibody IgG expression cassette for surface expression A schematic diagram showing pINT17-BSD, key features (1–7500 bp) is shown. pINT17-BSD-D1.3, a dual-promoter antibody expression cassette for surface expression of anti-lysozyme antibody D1.3. The complete annotated nucleic acid sequence is shown. Features: AAVS left homology arm 9~812 Blasticidin resistance gene 853~1254 pEF promoter 1522~2705 BM40 Leader 2745~2799 Humanized D1.3 VL 2799~3130 Human C Kappa 3138~3443 BGH Poly A 3468~3682 CMV promoter 3701~4273 Mouse VH leader with intron 4299-4426 Humanized D1.3 VH 4432-4779 Optimized human IgG1 CH1-CH3 4780~5775 Myc tag 5776~5805 PDGFR Anchor 5806~5961 BGH Poly A 6011~6225 AAVS right homology arm 6288~7124 f1 replication origin 7282~7695 pUC origin of replication 7916~8590 Kanamycin resistance gene 9310~10104 [Figure 2] Expression of IgG on the cell surface. Analysis was focused on viable cells using forward scatter and staining in the FL3 channel. Cells positive for staining in the FL3 channel (representing nonviable cells that had taken up 7-AAD) were excluded. HEK293 cells were transfected with pINT17 antibody in the presence of the AAVS TALEN. Stable populations were selected with blasticidin. 14 days after transfection, cells were stained with anti-Fc PE (FL2). Panels show fluorescence intensity (anti-Fc-PE, x-axis) plotted against cell number (y-axis) for the CNTO607 (a), MEDI1912 (b), and Ang2 mAb (c) pairs. For all panels, plots include stained HEK293 cells (dotted line), parental antibody (dashed line), and improved variants (solid black line). Figure 3: Preparative size-exclusion chromatography of MEDI-1912 and MEDI-1912-STT. Antibodies were expressed by transient transfection of Expi293 cells followed by Protein A affinity purification and dialysis. Purified MEDI-1912 (0.5 ml, 1.1 mg / ml) or MEDI-1912-STT (0.5 ml, 1.7 mg / ml) was loaded onto a Superdex 200 10 / 300 column connected to an AKTA Pure system using PBS (pH 7.4) running buffer. Elution volume (ml) is plotted on the x-axis against absorbance at 280 nm (mAU) on the y-axis. The elution volumes (Ve) of MEDI-1912 and MEDI-1912-STT were 10.3 ml and 11.7 ml, respectively. MEDI-1912 shows a prior elution volume, indicating self-interaction. [Figure 4] DNA and protein sequences of the MED-1912 variable heavy chain (VH). The primers employed for library generation are labeled on the nucleic acid sequences. [Figure 5] FACS separation of mixed cells displayed antibody populations based on antibody expression. An equal mix of MEDI-1912 and MEDI-1912_STT IgG genes was targeted to the AAVS locus of HEK293 cells via nuclease-directed integration. 15 days after transfection, this mixed cell population was separated based on antibody expression by FACS using a BD Influx sorter. Cells were stained with phycoerythrin (PE)-labeled anti-Fc and allophycocyanin (APC)-labeled NGF-biotin / streptavidin. Viable cells were analyzed using forward scatter and staining. Cells positive for staining in the λem=450 / 40, λexc=355 channels (representing nonviable cells that had taken up 7-AAD) were excluded. a. Histogram plot of anti-Fc-PE fluorescence intensity versus cell number for the mixed input population (dashed line), and monoclonal HEK293 cell lines displaying MEDI-1912 (gray line) and MEDI-1912_STT (black line). b. Dot plot shows the fluorescence intensity for anti-Fc-PE (x-axis), representing antibody expression levels, plotted against the fluorescence intensity for antigen binding (NGF-biotin / NGF-biotin / streptavidin-APC) on the y-axis. Gates selected to separate high and low antibody-expressing populations are labeled P5 and P6 and are shown as black boxes on the dot plot. The total number of events was 3.9 x 10 6 The number of cells sorted in the P5 and P6 gates was 2.5 x 10 5 and 2.8 × 10 5 It was an individual cell. [Figure 6] Enrichment of selected antibodies by mammalian cell display level. Cells sorted in gates 5 and 6 (Figure 5) were expanded, genomic DNA was prepared, and antibody VH genes were isolated by PCR. The two antibody populations were analyzed by Nextgen sequencing to determine the percentage of MEDI-1912 and MEDI-1912_STT in the two gated populations. Histograms show the percentage frequencies of MEDI-1912 (hatched bars) and MEDI-1912_STT (black bars) in the low antibody-expressing population (gate 6) and the high antibody-expressing population (gate 5). [Figure 7] FACS separation of MEDI-1912 library antibody populations based on antibody expression. A library of MENS-1912 IgG genes was targeted to the AAVS locus of HEK293 cells via nuclease-directed integration. NNS oligonucleotide-directed mutagenesis was used to randomly mutate the codons encoding W30, F31, and L56. Fifteen days after transfection, this mixed cell population was separated based on antibody expression and antigen binding by FACS using a BD Influx sorter. Cells were stained with phycoerythrin (PE)-labeled anti-Fc and allophycocyanin (APC)-labeled NGF-biotin / streptavidin. Viable cells were analyzed using forward scatter and staining. Cells positive for staining in the λem=450 / 40, λexc=355 channels (representing nonviable cells that had taken up 7-AAD) were excluded. Dot plots show the fluorescence intensity for anti-Fc-PE (x-axis), representing antibody expression levels, plotted against the fluorescence intensity for antigen binding (NGF-biotin / NGF-biotin / streptavidin-APC) on the y-axis for the parental MEDI-1912 monoclonal cell line (a), MEDI-1912_STT monoclonal cell line (b), and the MEDI-1912 amino acid 30, 31, and 56 random library (c). Gates selected for analysis are labeled P5 and P6 and are shown as boxes on the dot plots. [Figure 8] Sequence distribution of selected MEDI-1912 variants. Histogram plot of amino acid identity frequency at MEDI-1912 VH positions 30, 31, and 56 for the MEDI-1912 VH library after mammalian display, and FACS gated for Fc expression and NGF binding (P5 gate, Figure 7). Amino acids (single letter code) are plotted on the x-axis against frequency (percentage occurrence) on the y-axis for adjacent amino acids 30, hatched bar, and 31, black bar (a), and amino acid 56 (b). Leucine at position 56 was excluded from the analysis. [Figure 9] Alignment of bococizumab murine parent antibody 5A10 VH (A) and VL (B) with humanized intermediate antibody 5A10-i and bococizumab. The CDRs are indicated by bars above the sequence, and the mutated residues are highlighted in bold and underlined. Paratopic residues (amino acids that contribute to direct binding to PCSK9) are highlighted in italics and underlined. Dots indicate identity with the parental murine mAb 5A10. [Figure 10] Antibody mammals display expression of bococizumab and the parent humanized intermediate antibody 5A10-i The targeting vector pINT17 encoding bococizumab or 5A10-i IgG was integrated into the AAVS locus of Hek293 cells via TALE nuclease. At 1, 8, or 21 days post-transfection (dpt), cells (10 6 ) were stained with anti-Fc-PE, and 10 cells were analyzed by flow cytometry on an iQue Intellicyte flow cytometer. Dead cells were excluded from the analysis. Histogram plots show fluorescence intensity (anti-Fc-PE) versus cell number for bococizumab and 5A10-i cell display populations at 1, 8, and 21 days post-transfection (dpt) with staining of wild-type HEK293 cells incubated as a negative control. [Figure 11] Alignment of bococizumab VH with human germline sequence (IMGT). The bococizumab VH (Query_1) was subjected to an Ig Basic Local Alignment Search (IgBLAST) against the human VDJ database. Results are presented as an alignment to the query sequence (ordered by percent identity) (column 2), encompassing framework region 1 (FR1), complementarity-determining region 1 (CDR1), FR2, CDR2, and FR3. The human germline is shown in column 1. Residue identities to the bococizumab VH sequence (.) and differences (single amino acid codes) are displayed in the multiple alignment. [Figure 12] DNA sequence encoding the VH variants and VL of bococizumab plus a stop codon template. Variations compared to the original "wild-type" bococizumab (row f) are highlighted in bold and underlined. The adjacent 5' and 3' VH restriction sites (NcoI and XhoI) or VL restriction sites (NheI and NotI) are underlined. The VL stop codon is highlighted in bold and underlined. [Figure 13] Flow cytometry analysis of bococizumab library after MACS purification Hek293 cells transfected with the bococizumab library were MACS-purified with either anti-Fc or PCSK9i at 7 days post-MACS. (a) Flow cytometry dot plots are shown for the purified library, HEK293 cells, and unsorted library, bococizumab, and 5A10i transfectants, with anti-Fc expression (FL2, x-axis) plotted against PCSK9 binding (FL4, y-axis) for 9 days post-MACS. (b) Histograms of fluorescence intensity (anti-Fc, FL2, x-axis) plotted against cell number. Plots are (top to bottom) HEK293 control, bococizumab, 5A10i, bococizumab library, anti-PCSK9 MACS-purified bococizumab library, and anti-Fc MACS-purified bococizumab library. [Figure 14] BD Influx sorter dot plot of previously MACS-purified bococizumab library based on antigen binding or anti-Fc. A library of bococizumab IgG genes was targeted to the AAVS locus of HEK293 cells via nuclease-directed integration. This mixed cell population was first MACS-purified based on PCSK9 binding (a) or anti-Fc (b). Sixteen days after transfection, the MACS-enriched library was separated based on antibody expression and antigen binding by FACS using a BD Influx sorter. Cells were stained with phycoerythrin (PE)-labeled anti-Fc and allophycocyanin (APC)-labeled PCSK9-biotin / streptavidin. Analysis was focused for viable cells using forward scatter and staining. Cells positive for staining in the λem=450 / 40, λexc=355 channels (representing nonviable cells that had taken up 7-AAD) were excluded. The dot plot shows the fluorescence intensity for anti-Fc-PE (x-axis), which represents antibody expression levels, plotted against the fluorescence intensity for antigen binding (PCSK9-biotin / streptavidin-APC) on the y-axis. The gates selected for analysis are labeled P5 and P6, which are shown as boxes on the dot plot. [Figure 15] Bococizumab VH distribution after mammalian display selection. Random unselected input clones (84), sorted antigens (75), and selected Fc clones (85) were sequenced to determine VH identity. Histogram plots show VH germline identity on the x-axis plotted against percentage occurrence for input (white filled bars), antigen MACS followed by selected antigen and anti-Fc FACS (black filled bars), and anti-Fc MACS followed by selected antigen and anti-Fc FACS (gray filled bars) mammalian cell selection populations. [Figure 16] Bococizumab VL sequence analysis after mammalian display selection. Random unselected input clones (84), selected antigens (75), and selected Fc clones (85) were sequenced to determine the VL sequences. The average pI and aliphatic index were calculated for the three mutated codons, which showed a decrease in both pI and aliphatic index for the mammalian display-selected antibodies. [Figure 17] Table listing mammalian-displayed bococizumab clones enriched for antigen binding by MACS followed by FACS enrichment for both antibody display level (anti-Fc) and antigen binding. Clones were sequenced, and the single-letter amino acid sequences of VH CDR1 and CDR2 and VL CDR2 and CDR3 are shown, with variations from the original bococizumab sequence highlighted in bold and red. Targeted amino acids that retained the bococizumab sequence are underlined. Binding of antibodies, including the original parent antibody bococizumab and 5A10-i, to antigen in a capture ELISA was performed, and the binding signal in fluorescence units is shown in column 2. Liu et al., 2014 30 AC-SINS assays were performed as previously described by, and column 3 shows the maximum absorbance wavelength shift (nm) compared to a no-antibody PBS control. Selected human VH germlines are also lettered in column 6, as detailed in Example 5. a:VH Y33A-IGHV1-3*01 b:VH Y33D-IGHV1-8*01 c:VH S52N, F54S, R57S-IGHV1-46*01 d: VH Y33A, S52N, F54S, R57S (mutants a and c were combined) e: VH Y33D, S52N, F54S, R57S (variants b and c were combined) f: bococizumab "wild-type" sequence [Figure 18] HPLC-SEC of anti-PCSK9 IgG1 antibody. Antibodies were expressed by transient transfection of Expi-293 cells, affinity purified by Protein A chromatography, and dialyzed. Samples (2 μl at 1 mg / ml) were loaded onto an Agilent AdvancedBio SEC 300A, 2.7 μm, 4.6 x 300 mm column (Agilent Technologies, Cat. No. PL1580-5301) at a flow rate of 0.35 ml / min using an Agilent 1100 HPLC instrument. Absorbance versus retention time plots are shown for selected antibodies. Shades from black to progressively lighter gray are 5A10-i, 884_01_G01 (identified by mammalian cell display), bococizumab, and alirocumab. [Figure 19] Gel filtration analysis of nivolumab (a) and besencumab (b). Antibodies were expressed by transient transfection of Expi293 cells followed by Protein A affinity purification and dialysis. Purified nivolumab (0.5 ml, 1.3 mg / ml) or besencumab (0.5 ml, 1.2 mg / ml) was loaded onto a Superdex 200 10 / 300 column connected to an AKTA Pure system using PBS (pH 7.4) running buffer. Elution volume (ml) was plotted on the x-axis against absorbance at 280 nm (mAU) on the y-axis. The elution volumes (Ve) for nivolumab and besencumab were 12.0 ml and 13.7 ml, respectively. [Figure 20] Stability determination of nivolumab (a) and besencumab (b) after 2 weeks of storage at 4oC. Besencumab and nivolumab were purified by size-exclusion chromatography (see Figure 19) and adjusted to 0.5 mg / ml in PBS (pH 7.4). The antibodies were then stored at 4°C for 2 weeks. Dynamic light scattering measurements were performed at 20°C using a Zetasizer APS (Malvern Instruments, Malvern, UK) according to the manufacturer's instructions. The hydrodynamic radius was evaluated using the Einstein-Stokes equation and plotted against the scattering intensity. A single monodisperse peak was observed for nivolumab (a) compared to multiple aggregate peaks for besencumab (b). [Figure 21] Human serum binding to IgG on cell surfaces. Analysis was focused on viable cells using forward scatter and staining in the FL3 channel. Cells positive for staining in the FL3 channel (representing nonviable cells that had taken up 7-AAD) were excluded. Cells were transfected with pINT17-nivolumab or pINT17-besencumab in the presence of the AAVS TALEN. Stable populations were selected with blasticidin. Twenty days after transfection, cells were stained with human serum (H4522, Sigma) labeled with anti-Fc PE (FL2) and Dylight 633 (325-0000, Innova). Panels show untransfected HEK293 cells (a), HEK293 cells transfected with pINT17-nivolumab (b), or pINT17-besencumab (c). [Figure 22] Relationship between affinity, antigen concentration, and antibody concentration a K D Concentration of the complex using 0.1 nM antigen with different concentrations of antibody, either 10 nM (dashed line) or 0.1 nM (solid line). bi Lower affinity (K D Higher affinity antibodies (K 10 nM) D Relative selectivity of binding to 0.1 nM antigen for 0.1 nM Even at low antigen concentrations ("stringent"), there is relatively low selectivity at high antibody concentrations, but this increases as the antigen concentration decreases. [Figure 23] Splice acceptor / donor variants to control antibody display levels. The nucleic acid sequences from the HindIII site through the 5' intron, including the splice donor region, are shown for the pINT17-J9, J10, J29, and J30 variants. The original human "wild-type" sequence is J9, and nucleotides that differ from J9 at the splice junction for J10, J29, and J30 are underlined. [Figure 24] pINT17-J30, a dual promoter antibody IgG expression cassette for reduced display surface expression. The annotated nucleic acid sequence is shown between the XhoI (4804) and SbfI (8387) restriction sites. Features: IgG1 CH1-3 4805~5808 Splice Junction 5801~5802 Introns 5802-7104 M1 exon 7105–7239 BGH pA 7264~7478 AAVS right homology arm 7544~8381 3'β-globin insulator 8421~8492 [Figure 25] Reduced surface expression of IgG on the cell surface using alternative transmembrane domains and splice variants. Analysis was focused on viable cells using forward scatter and staining in the FL3 channel. Cells positive for staining in the FL3 channel (representing nonviable cells that had taken up 7-AAD) were excluded. Cells were transfected with the pINT17 targeting vector in the presence of the AAVS TALEN. Stable populations were selected with blasticidin. 27 days after transfection, cells were stained with anti-Fc PE (FL2). Flow cytometry dot plot panels include pINT17-J9-nivolumab (a), pINT17-J10-nivolumab (b), pINT17-J29-nivolumab (c), pINT17-J30-nivolumab (d), and pINT17-BSD-nivolumab (e). [Figure 26] Quantification of IgG display levels on the cell surface for antibodies expressed from pINT17-BSD or pINT17-J30 targeting vectors Calibration bead FL2 staining was performed as described in the manufacturer's instructions for Quantum Simply Cellular anti-mouse IgG beads (catalog no. 815, Bangs Laboratories Inc.) stained with mouse IgG-PE conjugate. (a) Labeled histogram plot shows staining of the calibration bead set with peaks labeled 1, 2, 3, and 4, representing bead copy numbers of 12257, 72745, 283360, and 886417, respectively. The blank peak represents beads without capture antibody. (b) Calibration graph showing median fluorescence intensity (x-axis) plotted against copy number (y-axis). Cell lines expressing 337_1_C08 (c) and nivolumab (d) from either the pINT17-BSD or pINT17-J30 expression cassette were stained with anti-Fc-PE (5 μl, 0.1 mg / ml; 10 5 Cells were stained with 7-AAD (cells). Analysis was focused on viable cells using forward scatter and staining in the FL3 channel. Cells positive for staining in the FL3 channel (representing nonviable cells that had taken up 7-AAD) were excluded. Histogram plots show fluorescence intensity versus cell number for pINT17-BSD (labeled and black line), pINT17-J30 (labeled and dotted line) with PDGFR TM, and wild-type HEK293 cell line (gray solid line) for cells expressing 337_1_C08 (c) and nivolumab (d). [Figure 27] Copy number reduction in cell display allows the separation of antibodies with different affinities for their targets by mammalian display. Hek293 cells displaying nivolumab and 337_1_C08 antibodies were labeled with 50 nM Cell Tracker Green and 50 nM Cell Tracker Red, respectively. (a) Demonstrates display using the J30 splice variant, and (b) demonstrates display using the PDGFR transmembrane domain encoded by the pINT17-BSD vector. Labeled cells were mixed equally and MACS sorted based on antigen binding. Sorted cells were analyzed using an Intellicyt flow cytometer. Dot plots represent nivolumab on the x-axis (FL1) and 337_1_C08 on the y-axis (FL4). Panels i, ii, iii, and iv represent 10 nM, 1 nM, 0.1 nM, and no antigen employed for MACS purification, respectively. [Figure 28] pINT18-Tet1, an inducible promoter antibody IgG expression vector for reduced display surface expression. The annotated nucleic acid sequence is shown between the AsiSI (5) and SbfI (7672) restriction sites. The vector backbone outside the AsiSI and SbfI sites (7673-10922 and 1-4), including the origin of replication and kanamycin resistance gene, is identical to pINT17-BSD (Figure 1). Key Features: AAVS left homology arm 9~812 Blasticidin resistance gene 853~1254 CMV promoter 1540~2112 Reverse Tet activator (tTA) CDS 2164~3168 SV40 pA 3178-3395 tetO heptamer 3679~3932 Minimal CMV promoter (PminCMV) 3946-4005 BM40 Leader 4016~4066 Anti-PD1 MK3475 VL 4068~4413 Human C-kappa 4421~4738 Furin cleavage site 4745~4756 P2A peptides 4757-4816 Mouse VH leader with intron 4829-4960 Anti-PD1 MK3475 VH 4962~5321 Optimized human IgG1 CH1-CH3 5322~6317 Myc tag 6318~6347 PDGFR Anchor 6348~6503 BGH Poly A 6553~6767 AAVS right homology arm 6829~7666 3'β-globin insulator 7706~7777 f1 replication origin 7824~8237 pUC origin of replication 8458~9132 Kanamycin resistance gene 9852~10646 [Figure 29] Inducible mammalian display expression Histograms showing staining results from a HEK293 cell line co-transfected with pINT18-Tet1-377_1_C08 and TALE nuclease and a stable cell population selected for 20 days in the presence of blasticidin. Samples were collected at 5 × 10 5 The cells were split into 100 cells / ml and induced with 20 ng / ml, 2 ng / ml, and 0 ng / ml doxycycline. 24 hours after induction, 1 × 10 cells from each doxycycline-induced sample were cultured. 6 Flow staining was performed using cells. Cells were stained using anti-Fc-PE and TOPRO-3 viability stain. The histogram shows the fluorescence intensity for the FL2 channel (anti-Fc-PE) plotted against cell number. HEK293 WT control (gray solid line), HEK293-pINT18-Tet1-377_1_C08 stable cell line induced with 0 ng / ml doxycycline (black dashed line), 2 ng / ml doxycycline (black dotted line), and 20 ng / ml doxycycline (black solid line). [Figure 30] Binding of cells displayed antibodies to FcRn. HEK293 cells expressing Briakinumab and Ustenkinumab were stained with biotinylated FcRn (50 nM) pre-complexed with streptavidin PE (11 nM) using different buffers. a.Hek293 WT b. Streptavidin PE control c. Cells stained with buffer pH 6.0 d. Cells stained with buffer pH 7.4 [Figure 31] FACS separation of HEK293 cells displayed levels of anti-mesothelin IgG. A population of anti-mesothelin antibody genes was integrated into the human AAVS locus of HEK293 cells by nuclease-mediated gene transfer. Polyclonal populations of antibody-expressing HEK293 cells were separated by FACS according to antibody display level by staining with anti-human Fc-PE. Sixteen days after transfection, the MACS-enriched library was separated based on antibody expression by FACS using a BD Influx sorter. Cells were stained with anti-Fc conjugated with phycoerythrin (PE). Analysis was focused for viable cells using forward scatter and staining. Cells positive for staining in the λem=450 / 40, λexc=355 channels (representing nonviable cells that had taken up DAPI) were excluded. The histogram shows the fluorescence intensity for anti-Fc-PE (x-axis), representing antibody expression level, plotted against cell number on the y-axis. The gates selected for analysis are labeled P4, P6, and P5, representing populations with low, medium, and high display levels, respectively. [Figure 32] HPLC-SEC of two anti-mesothelin IgG1 antibody clones derived from the high display level group (solid line) and the low display level group (dotted line), respectively. Figure 33: DNA binding and depletion of DNA-binders using MACS. (A) Overlay of HEK293 cells displaying ustekinumab (dashed line), briakinumab (long-dashed line), and amatuximab (dotted line) stained with HEK293 cells (solid gray line) or biotinylated DNA detected with streptavidin PE. (B) Dot plots representing the mixture of three antibody-cell populations displaying ustekinumab (unlabeled, Q4), amatuximab (labeled with CellTace Far red, X-axis), and briakinumab (labeled with CellTrace CFSE, Y-axis) stained with DNA prior to MACS sorting. (C) Dot plots of the flow-through showing depletion of DNA-binders. [Figure 34] Double staining with heparin-FITC (x-axis) and anti-human Fc APC (y-axis). (a) Dot plot showing an overlay of ustekinumab (gray) and briakinumab (black). (b) Dot plot showing an overlay of ustekinumab (gray) and ganitumab (black). Gates within the overlay plots indicate cells that are high expressers and non-binders of heparin. Figure 35: Double staining with chaperones conjugated with DyLight 633 (x-axis) and anti-human Fc PE (y-axis). (a) Dot plot showing an overlay of ustekinumab (gray) and briakinumab (black) double stained with Hsp70-DyLight 633 and anti-human Fc PE. (b) Dot plot showing an overlay of ustekinumab (gray) and briakinumab (black) double stained with Hsp90-DyLight 633 and anti-human Fc PE. Gates within the overlay plot indicate cells that are high expressers and non-binders of chaperones (Hsp70 and Hsp90). (c and d) Overlay histogram plots show the binding of lenzilumab and brentuximab to Hsp70 and Hsp90, respectively. Figure 36: Histogram plots of antibodies and polyreactive probes stained with anti-human Fc PE. Stable monoclonal HEK293 cell lines displaying a selection of antibodies were created by nuclease-mediated transfection. Histogram plots of fluorescence intensity (x-axis) versus cell number (y-axis) are shown for various antibodies displayed on the surface of HEK293 cells with the following probes: (a) anti-human Fc-PE, (b) biotinylated DNA detected using streptavidin PE, (c) heparin-FITC, (d) streptavidin PE, (e) Hsp70-DyLight 633, (f) Hsp90-DyLight 633, and (g) FcRn precomplexed with streptavidin PE. [Figure 37] pINT17-Tet-D1.3, an inducible antibody IgG mammalian display expression vector. The complete annotated nucleic acid sequence is shown between the AAVS homology arms and the promoterless blasticidin gene from the BglII to BstZ171 restriction sites. Numbering is from the BglII restriction site. Key features are listed below. BGH polyA 223~9 (reverse strand) Human C kappa 544~236 (reverse chain) D1.3 VL 877~549 (reverse chain) Human VL leader with intron 1168-883 (opposite strand) TRE3G promoter 1230~1618 CMV promoter 1237~1809 VH leader with intron 1644-1782 D1.3 VH 1783~2127 IgG1 CH1-3 2125~3120 Myc tag 3121~3150 PDGFR Anchor 3151~3306 BGH Poly A 3356~3570 pEF promoter 3621~4955 rtTA-3G 5063~5809 SV40 PolyA 5832~6274 [Figure 38] Inducible IgG mammalian display cell lines. 1549_02_D06 (1), 1535_01_E03 (2), and 337_1_C08 (3), bococizumab (4), 884_01_G01 (5), 5A10i (6), and alirocumab (7). At 27 dpt, cell lines were induced by the addition of 0 (a), 2 (b), 4 (c), or 100 (d) ng / ml doxycycline. 24 hours after induction, cells were stained with anti-Fc-PE. Histogram of fluorescence intensity (anti-Fc, FL2, x-axis) plotted against cell number. [Figure 39] Inducible IgG mammalian display cell line: cell surface IgG turnover pINT17-Tet:1549_02_D06 (1), 1535_01_E03 (2), and 337_1_C08 (3), containing the VH and VL of anti-PD1 antibodies, and the anti-PCSK9 antibodies bococizumab (4), 884_01_G01 (5), 5A10i (6), and alirocumab (7), were used to generate stable HEK293 cell lines by AAVS TALE nuclease-mediated gene integration and blasticidin selection. At 27 days post-induction, cell lines were induced by the addition of 100 ng / ml doxycycline. Forty-eight hours after induction, cells were stained with anti-Fc-PE. Histograms of fluorescence intensity (anti-Fc, FL2, x-axis) are shown plotted against cell number. [Figure 40] Anti-PD1 antibody 1549_02_D06 (K for PD-1) D = 2.9 nM) and 337_1_C08 (K for PD-1 DCell lines displaying PD-1-biotin (p = 74 nM) were induced with (a) 0, (b) 2, (c) 4, and (d) 100 ng / ml doxycycline, respectively. Dot plots of fluorescence in the FL1 channel (y-axis) against forward / side scatter (FSC, x-axis) are shown. Labeled cells displaying 1549_02_D06 are shown in the upper quadrant of each dot plot, and unlabeled cells displaying 337_1_C08 are shown in the lower quadrant. Panels i, ii, iii, and iv represent 0.1, 1, and 10 nM concentrations of PD-1-biotin employed for MACS purification, respectively. Panel iv represents the input pre-MACS population. The percentage of each cell population is indicated within each quadrant. Figure 41: Overlay dot plot of double-stained populations of ustekinumab (gray) and briakinumab (black). Double staining with 50 nM FcRn-Avi tag pre-complexed with streptavidin PE (x-axis) and anti-human Fc APC (y-axis). The gate within the plot represents ustekinumab as an FcRn non-binder that can be sorted by FACS from FcRn binders. Figure 42: Germline analysis of anti-mesothelin variable heavy (VH) domain antibody populations. The figure plots the frequency of occurrence for each VH germline in input and low, medium, and high mammalian display gated populations. Figure 43: Germline analysis of anti-mesothelin variable light kappa (VLκ) domain antibody populations. The figure plots the frequency of occurrence for each VLκ germline in input and low, medium, and high mammalian display gated populations. Figure 44: Germline analysis of anti-mesothelin variable light lambda (VLλ) domain antibody populations. The figure plots the frequency of occurrence for each VLλ germline in input and low, medium, and high mammalian display gated populations. [Figure 45] HPLC-SEC of anti-mesothelin IgG1 antibody. The antibody was expressed by transient transfection of Expi-293 cells, affinity-purified by Protein A chromatography, and dialyzed. The sample (2 μl at 1 mg / ml) was loaded onto an Agilent AdvancedBio SEC 300A, 2.7 μm, 4.6 × 300 mm column (Agilent Technologies, Cat. No. PL1580-5301) at a flow rate of 0.35 ml / min using an Agilent 1100 HPLC instrument. Plots of absorbance at 215 nm against retention time are shown for selected anti-mesothelin antibodies: 932_01_A03 (black line) and 930_01_A12 (alternating dotted and dashed lines), 930_01_B02 (long dashed line), from the high display level group, and 930_01_C12 (short dashed line), from the low display level group. Figure 46: Alignment of human and CHO AAVS intron 1 TALE-nuclease (TALEN) target binding sites. The DNA sequence of CHO AAVS intron 1 was obtained from the ENSEMBL-annotated CHO-K1 glutamine synthetase (GS) knockout cell line, accession number: CHOK1GS_HDv1:scaffold_52:2374828:2406177:1. Numbering refers to the start of human PPP1R12C intron 1. Bold text indicates the left and right arms of the human TALEN target site. Asterisks indicate homology between the human and CHO sequences, and dashed lines (-) indicate deletions. Underlining and italics indicate the ends of the AAVS left and right homology arms in the pINT17 targeting vector. This alignment was used to design the CHO AAVS homology arms in the pINT17-CHO targeting vector and, for comparison, CRISPR / Cas9 guide RNAs. Sense or antisense CRISPR guide RNA recognition sites, numbered 1–3, are indicated above or below the sequence, respectively. [Figure 47] CHO AAVS homology arms in vector pINT17-BSD-CHO, a dual promoter antibody IgG expression cassette for surface expression on CHO cells. Annotated DNA sequences are shown for the left and right CHO AAVS homology arms in the vector. All remaining features encompassing the dual promoter antibody expression cassette, including those not shown in this figure, are as described for vector pINT17-BSD (Figure 1) and are listed below. Features: CHO AAVS left homology arm 9~899 Blasticidin resistance gene 942~1343 pEF promoter 1611~2794 BM40 Leader 2834~2885 Humanized D1.3 VL 2888~3219 Human C Kappa 3227~3532 BGH Poly A 3468~3682 CMV promoter 3790~4362 Mouse VH leader with intron 4388-4515 Humanized D1.3 VH 4521~4868 Optimized human IgG1 CH1-CH3 4869~5864 Myc tag 5865~5894 PDGFR Anchor 5895~6050 BGH Poly A 6100~6314 CHO AAVS left homology arm 6376~7266 f1 replication origin 7424~7837 pUC origin of replication 8058~8732 Kanamycin resistance gene 9452~10246 [Figure 48] Display of antibodies on the surface of CHO cells by TALEN or CRISPR / Cas9 nuclease-mediated gene integration. Histogram of fluorescence intensity (anti-Fc, FL2, x-axis) plotted against cell number. Plots (top to bottom) are: CHO control, pINT17-BSD-CHO V2-nivolumab minus nuclease, pINT17-BSD-CHO V1-nivolumab minus nuclease, pINT17-BSD-CHO V1-nivolumab plus CHO TALEN, pINT17-BSD-CHO V1-nivolumab plus CRISPR3, pINT17-BSD-CHO V1-nivolumab plus CRISPR2, pINT17-BSD-CHO V1-nivolumab plus CRISPR1. [Figure 49] Antibody display levels on the surface of CHO. Histograms of fluorescence intensity (anti-Fc, FL2, x-axis) plotted against cell number (filled plot) of CHO cells (a) bococizumab (solid line) and 884_01_G01 (dashed line). (b) MEDI-1912 (solid line) and MEDI-1912-STT (dashed line). [Figure 50] Creating "enhanced developability" populations using mammalian display for subsequent binding selection a. Anti-PD1 337_1_C08 VH(i) and VL(ii) chain sequences. Nucleotide sequences are shown with the translated single-letter amino acid code above the codons. CDRs are annotated (underlined), and CDR3 amino acids targeted for site-directed mutagenesis are highlighted in bold. b. Anti-PD1 antibody VH and VL CDR3 mammalian display libraries were separated by FACS based on high, intermediate, and low antibody cell display levels and analyzed by flow cytometry after staining with anti-Fc-PE. Histogram plots of cell number (y-axis) versus Fc expression (x-axis) are shown (top to bottom) as high (i), intermediate (ii), and low (iii) anti-PD1 populations. For reference, the anti-Fc MACS population (iv), the "wild-type" 337_1_C08 parental clone (v), and HEK293 cells without displayed antibody (vi) are shown. [Figure 51] pINT17-Bi-CMV-Emicizumab, a plasmid for cell surface expression of the bispecific "knobs-into-holes" containing bidirectional CMV and elongation factor (pEF) promoter, common light chain IgG emicizumab. This is a tricistronic targeting vector with three promoters driving the expression of three genes: anti-FIXa heavy chain, anti-FX heavy chain, and common light chain. The complete annotated nucleic acid sequence is shown between the AAVS homology arms from the BglII to BstZ171 restriction sites. Numbering is from the BglII restriction site. Key features are listed below. BGH polyA 222~8 (reverse strand) Human C kappa 546-232 (reverse chain) Emicizumab VL 876~547 (reverse chain) Human VL leader with intron 1143-884 (opposite strand) Minimal CMV promoter 1230-1167 (reverse strand) CMV promoter 1237~1809 Mouse VH leader with intron 1835-1973 Emicizumab anti-FIXa VH 1974~2339 Emicizumab anti-FIXa CH1-3 2340~3317 Myc tag 3318~3347 PDGFR Anchor 3348~3503 BGH Poly A 3553~3767 pEF promoter 3818~5152 Human VH leader with intron 5260-5401 Emicizumab anti-FX VH 5402~5758 Emicizumab anti-FX CH1-3 5759~6733 Myc tag 6734~6763 PDGFR Anchor 6764~6916 SV40 PolyA 6942~7384 [Figure 52] Binding of FIXa and FX displayed on the surface of HEK293 cells to the bispecific antibody emicizumab pINT17-Bi-CMV-emicizumab or pINT17-BSD-anti-FIXa was used to transfect HEK293 cells in the presence of a plasmid encoding the AAVS TALEN. 24 hours after transfection, cells were analyzed for antibody display and their ability to bind the antigens FIXa or FX. Histogram plots show, from left to right, the number of cells versus fluorescence intensity when stained with: (a) bispecific emicizumab (dashed black line), (b) anti-FIXa IgG (solid black line), (c) HEK293 cells displaying anti-Fc-APC, FX-biotin or FIXa-biotin pre-complexed with streptavidin-PE, or streptavidin-PE alone. [Figure 53] Alignment of emicizumab VL with parent emicizumab VL. The CDRs are indicated by bars above the sequences. Dots indicate identity with the final emicizumab VL. Residues contributing to the positive charge patch are highlighted in bold. [Figure 54] Display of knotbodies on the surface of HEK293 cells and their relationship to their biophysical properties. (a) HEK293 cells were transfected with pINT17-knotbody in the presence of the AAVS TALEN. Stable populations were selected with blasticidin. Seven days after transfection, cells were stained with anti-Fc PE (FL2) and analyzed by flow cytometry. The histogram shows a plot of cell number (y-axis) against fluorescence intensity (anti-Fc-PE, x-axis) for KB_A12 EETI-II (solid black line), KB_A12 Hstx1 (dotted line), and KB_A12 ProTxIII (dashed line). The traces of KB_A12 Hstx1 (dotted line) and KB_A12 ProTxIII (dashed line) overlap and appear to merge. Knotbodies were expressed by transient transfection of Expi293 cells and purified by Protein A affinity chromatography. Knotbodies were analyzed by HPLC-SEC as described above, and plots of absorbance versus elution volume are shown for (b) KB_A12 EETI-II, (c) trastuzumab, and (d) KB_A12 ProTx-III. Figure 55. Knotbody mutant libraries contain clones with improved display levels compared to the parent knotbody. HEK293 cells displaying knotbodies were stained with anti-Fc-PE and analyzed by flow cytometry. Histogram plots of cell number versus fluorescence intensity are shown for three libraries (after anti-Fc MACS purification) compared to the relevant parent knobody control displaying cell line. (From left to right): (a) KB_A12 ProTxIII Library Set A (dotted line) and KB_A12 ProTxIII control (black line), (b) KB_A12 ProTxIII Library Set B (dotted line) and KB_A12 ProTxIII control (black line), (c) KB_A12 HsTx1 Library (dotted line) and KB_A12 HsTx1 control (black line). [Example]

[0360] Example 1. Construction of targeting vectors for soluble expression and cell surface displayed antibodies in the IgG format To enable the display of binder molecules, including antibodies, on the surface of higher eukaryotic cells and their subsequent genetic selection, vectors can be used to target binder genes to specific locations in the host genome. The vectors can encode selectable markers to enable the selection of stable cell lines; the selectable markers can encode genes that confer resistance to blasticidin, G418 / geneticin, hygromycin, puromycin, or zeocin. The targeting vectors can contain exogenous promoters that drive expression of the selectable marker-encoding gene. Alternatively, the transgene can be integrated into cellular DNA downstream of an endogenous promoter, allowing preferential selection of correctly integrated transgenes. The targeting vectors will also encode homology arms to enable homologous recombination into the relevant chromosomal locus and promoter, driving expression of the binder molecule and polyadenylation (pA) site. The binding agent molecule gene will be fused to DNA encoding a leader sequence that allows secretion through the endoplasmic reticulum (ER) to the cell surface and a membrane anchor such as a transmembrane domain or glycosylphosphatidylinositol (GPI) anchor.

[0361] A schematic of the targeting vector used herein is shown in Figure 1a, and the complete, annotated DNA sequence is shown in Figure 1b. The plasmid contains AAVS homology arms flanking the expression cassette, allowing homologous recombination of the transgene into the human AAVS locus. The AAVS locus was first identified as a common integration site for adeno-associated viruses and is a "safe harbor" locus for the insertion and expression of heterologous genes in human cells. 93 After nuclease-mediated cleavage within the AAVS site, the AAVS homology arms in the targeting vector facilitate integration of the expression cassette by homologous recombination. The blasticidin gene lacks a promoter in the vector but is preceded by a splice acceptor that creates an in-frame fusion with the upstream exon of the AAVS locus. Details of the antibody heavy and light chain expression cassettes are described below and in Figure 1.

[0362] The targeting vector pINT17-BSD (Figures 1a and 1b) was constructed by polymerase chain reaction (PCR) amplification of selected fragments from a previously described vector (WO2015166272A2) containing restriction sites that allow subsequent assembly. The origins of the various elements of pINT17-BSD are described below. DNA encoding the AAVS left homology arm, splice acceptor, blasticidin resistance gene, polyadenylation site, and elongation factor 1 alpha promoter (pEF1α) was derived from the pD2 plasmid (WO2015166272A2) by PCR amplification (1511 bp) with the addition of 5' AsiSI and 3' BglII restriction enzymes. DNA encoding the Myc tag and PDGFR transmembrane domain was PCR amplified from the pD2 plasmid with the addition of a 5' IgG1 CH3 homology sequence and a 3' HindIII site. The light chain BM40 leader and the variable light chain (VL) of the anti-lysozyme antibody D1.3 were also amplified. 94DNA encoding human constant light (CL), bovine growth hormone (BGH) pA, the immediate-early cytomegalovirus promoter (CMV promoter), the mouse heavy chain leader split by an intron, the variable heavy chain (VH) of the anti-lysozyme antibody D1.3, and IgG1 antibody constant heavy domains 1 to 3 (IgG1 CH1-3) were PCR amplified from the previously described pINT3 plasmid (WO2015166272A2) with a 5' BglII site and a 3' addition of DNA encoding a Myc tag. Two fragments encoding the 4446 bp region of pINT17-BSD from BglII to HindIII were combined by PCR assembly to add the PDGR transmembrane domain directly to the CH3 end. The region from HindIII to SbfI encoding the AAVS right homology arm was PCR amplified (1168 bp) from the pD2 plasmid (WO2015166272A2) with the addition of HindIII and SbfI restriction sites. The vector backbone includes the f1 and pUC origins of replication and the kanamycin resistance gene from pSF-EF1alpha (Oxford Genetics OG43) through the SbfI to AsiSI sites. The example shown in Figure lb encodes the VL and VH of a human anti-lysozyme antibody, but this can easily be substituted for other specificities using standard molecular biology techniques (e.g., replacing the VL and VH genes using flanking restriction enzymes).

[0363] Example 2. Comparison of surface display levels of parental and enhanced developable clones for three pairs of antibodies Three antibody pairs were investigated, where the original parent antibody retains a poor developability profile and their re-engineered daughter molecules have been modified to improve their self- and cross-interaction properties. The panel consisted of CNTO607, a monoclonal antibody against interleukin IL-13, and its modified counterpart, CNTO607 W100A. 14 CNTO607 was poorly soluble at neutral pH and precipitated at high concentrations in PBS buffer, leading to a poor solubility in the affinity capture self-interacting nanoparticle spectroscopy (AC-SINS) assay. 39The structure determination of CNTO697 revealed a hydrophobic patch in the heavy chain CDR3. The VH CDR3 mutation W100A improved both the antibody solubility and cross-interaction chromatography (CIC) profile. 47 CIC measures binding to human serum polyclonal antibodies immobilized on a column matrix. A second example is a monoclonal antibody named Ang2mAb that targets angiopoietin 2, a soluble ligand for the Tie2 receptor and a regulator of pathological angiogenesis. However, Ang2mAb has been reported to exhibit both poor expression and aggregation. A combination of structural modeling and experimental screening of 19 mutants identified the better-expressing Ang2mAb C49T, in which the unpaired cysteine ​​residue was mutated. 6 Finally, we included MEDI-1912, an anti-nerve growth factor (NGF) antibody that inhibits signaling through the TrkA and p75 receptors. 7 MEDI-1912 could potentially be used to treat chronic pain, but it exhibits precipitation and aggregation in solution and a poor pharmacokinetic profile. MEDI-1912 was affinity matured from a "grandparent" antibody named MEDI-578, which bound NGF with picomolar affinity and behaved well in terms of self-aggregation. Hydrogen / deuterium exchange-mass spectrometry (HDX-MS) and molecular modeling identified a hydrophobic patch on the VH domain caused by residues within VH CDR1 and CDR2. This allowed prediction of the amino acids responsible for self-association and resulting aggregation. This then led to the identification of NGF. 7 This allowed for the design of a triple mutant (MEDI-1912_STT) with mutations W30S, F31T, and L56T that disrupts the self-interaction interface while retaining potency and affinity for ATP.

[0364] Synthetic DNA encoding the heavy and light variable domains of CNTO607, CNTO607-W100A, Ang2mAb, Ang2mAb-C49T, MEDI-1912, and MEDI-1912_STT (see Table 1 for sequences) was cloned into the mammalian display vector pINT17-BSD (see Example 1 for vector map and sequence), the DNA sequence was confirmed, and transfection-quality plasmid DNA was prepared. One day prior to transfection, suspension-adapted HEK293 cells were cultured at 5 x 10 in HEK FreeStyle 293 expression medium. 5 Cells were seeded at 1 × 10 cells / ml in 10 ml. 6 PEI transfection was performed when the density of cells / ml was reached. The antibody gene carrying pINT17 (1 μg) and the left and right TALEN plasmids (5 μg each) were mixed and diluted in unsupplemented HEK FreeStyle 293 expression medium (1 ml). Polyethylenimine (PEI), linear, 25000 Da MW (10 μl, 1 mg / ml, Polysciences) was added and incubated at room temperature for 10 minutes. The plasmid DNA / PEI mixture was then added to HEK293 suspension cells (1 × 10 in 10 ml of HEK FreeStyle 293 expression medium). 6Blasticidin selection was initiated at a concentration of 7 μg / ml 48 hours after transfection. The population was kept under selection for the duration of the experiment. 15 days after transfection (dpt), cells were stained with anti-human Fc PE (Biolegend). Monoclonal cell lines expressing the antibodies were then stained using the following protocol: HEK293 cell lines expressing the antibodies or wild-type HEK293 cells (1 million cells) were pelleted (200 g for 3 minutes in an Eppendorf tube (1.5 ml)). The pellet was resuspended in PBS (1 ml) and centrifuged (600 g for 2.5 minutes). The pellet was resuspended in PBS (100 μl) containing 1% BSA and anti-Fc PE (5 μl, Biolegend). The mixture was incubated at 4°C for 30 minutes in the dark. 900 μl of 0.1% BSA in PBS was added, and the cells were pelleted (600 g, 2.5 min). The cells were resuspended in 1 ml of 0.1% BSA in PBS, and this wash step was repeated once. The cells were resuspended in 200 μl of 0.1% BSA in PBS containing 7-AAD (5 μl per million cells). 50 μl of labeled cells were analyzed using the Intellicyte iQue screener. Flow cytometry analysis (Figure 2) showed increased antibody display levels for the improved daughter molecules for all three antibody pairs compared to the original problematic parent molecules.

[0365] [Table 1]

[0366] Example 3a. Relationship between cell surface display levels and self-interaction at high concentrations Antibody expression and purification was performed to investigate the properties of the antibodies described in Example 2. Synthetic DNA encoding the heavy and light variable domains of CNTO607, CNTO607-W100A, Ang2mAb, Ang2mAb-C49T, MEDI-1912, and MEDI-1912_STT (see Table 1 for sequences) was cloned into a pINT3 (WO2015 / 166272A2)-based dual-promoter IgG soluble expression vector, and correct cloning was confirmed by DNA sequencing.

[0367] Plasmid DNA was prepared and used to transfect Expi293 cells (final culture volume: 30 ml) using the transfection reagent ExpiFectamine according to the manufacturer's instructions (A14525, ThermoFisher Scientific). 24 h prior to transfection, cells were seeded at a density of 2 x 10 cells / ml in 25.5 ml of Expi293 expression medium. Plasmid DNA (30 μg) was diluted in 1.5 ml of Opti-MEM medium, and 80 μl of ExpiFectamine 293 reagent was added to 1.5 ml of Opti-MEM medium and incubated at room temperature for 5 minutes. The diluted plasmid DNA (30 μg in 1.5 ml of Opti-MEM medium) was then added to the diluted ExpiFectamine 293 reagent (80 μl of ExpiFectamine in 1.5 ml of Opti-MEM medium) and incubated at room temperature for 20 minutes. Cells were incubated at 37°C, 5% CO2, 5% humidity and agitated at 130 rpm (25 mm orbital throw, ISF1-X, Climo-Shaker, Kuhner). After 5 days of expression, culture supernatants were harvested by centrifugation (2000g, 20 min).

[0368] The pH of the culture supernatant in a 50 ml centrifuge tube was adjusted by adding 1 / 10 volume of PBS (pH 7.4), and Protein A Sepharose FF resin (300 μl, Generon, PC-A100) was added and incubated with agitation at room temperature for 1 hour. The 50 ml tube was centrifuged at 2000 g for 5 minutes to recover the beads, and the supernatant was discarded, leaving approximately 1 ml of bead slurry. This slurry was loaded onto a fritted column (Proteus "1-Step Batch" midi centrifugal column, Generon, GEN-1SB08), centrifuged (50 g, 4°C for 1 minute), and the flow-through was discarded. The column was washed 2x with PBS (2 x 10 ml), followed by centrifugation (50 g, 4°C for 1 minute) after each wash step. The antibody was eluted using elution buffer (900 μl, 0.2 M glycine pH 2.6) added to the column matrix, and the eluate was immediately neutralized using neutralization buffer (300 μl, 1 M Tris-HCl, pH 8). The antibody was then eluted from the Protein A Sepharose column directly into the neutralization buffer by centrifugation (50 g, 1 min at 4°C). The antibody was buffer exchanged by transfer to a GeBAflex maxi tube (8 kDa molecular weight cutoff, Generon, D045) and dialyzed against 4 liters of PBS and incubated at 4°C for at least 3–18 hours. This dialysis step was repeated with a second 4 L PBS dialysis step. The antibody yield and concentration were determined by measuring absorbance at 280 nm and calculating using the Beer-Lambert law with an estimated extinction coefficient of 1.4 to estimate the concentration.

[0369] The yield of polypeptides produced by transient expression can be considered an indicator of developability. Comparison of expression yields in transient transfections between the three pairs of antibodies from Example 2 showed lower expression of the parental antibodies, but significant differences in developability potential were not evident simply by comparing the yields of transient transfections (Table 2). For example, the expression yield of the parental CNTO607 antibody was 34 mg / L, while the expression yield of the improved solubility CNTO607-W100A antibody was 55 mg / L. Similarly, the expression yield of the parental antibody MEDI-1912 was 33 mg / L compared to 53 mg / L for the improved version. The yield of Ang2 mAb was 13 mg / L compared to 34 mg / L for the engineered progeny Ang C49T.

[0370] The melting temperature of a polypeptide may be considered a surrogate for predicting "developability," and in some cases, antibodies may be more developable. 9~11 These antibodies were selected for their improved melting temperatures in the hopes of generating antibodies. Melting temperatures (Tm) and aggregation onset temperatures (Tagg) were determined using a Prometheus NT.4B (Nanotemper) according to the manufacturer's instructions. Approximately 8–10 μl of antibody solution at 0.5 mg / mL was collected using a small capillary. The capillary was then clipped in place and a fluorescence scan was performed with the Prometheus instrument for thermal melting analysis. The melting and scattering onset temperatures were determined using Prometheus fitting software. The antibody melting temperatures (Tm) and aggregation temperatures (Tagg) were similar both within the antibody pair set and compared to a clinically approved positive control anti-PD1 antibody (see Table 2). This data indicates that the melting and aggregation temperatures of this antibody set are not predictive of biophysical profiles of self-interactions and nonspecific cross-interactions.

[0371] During preparative size-exclusion chromatography, MEDI-1912 displayed a faster elution profile compared to MEDI-1912_STT (Figure 3), indicating that it exists as a higher molecular weight species and is prone to self-interaction. The remaining antibody eluted with a profile similar to that of nivolumab. To enable measurement of antibody self-interactions by dynamic light scattering (DLS), the size-purified antibodies were concentrated by ultrafiltration. The antibody concentrations achieved for each antibody pair are shown in Table 2. This revealed that it was impossible to concentrate the parent antibodies MEDI-1912 and CNTO607 beyond 1.4 mg / ml and 1.8 mg / ml, respectively, before antibody precipitation blocked the ultrafiltration membrane. In contrast, it was possible to concentrate the solubility-enhanced daughter molecules MEDI-1912_STT and CNTO607_W100A to 29 and 30 mg / ml, respectively, without evidence of precipitation. No precipitation was observed for the concentrated Ang2 mAb pair. Dynamic light scattering (DLS) detected higher-order aggregated species for the parent antibodies MEDI-1912 and CNTO607 (Table 2), but not for the daughter molecules MEDI-1912_STT and CNTO607_W100A, as judged by their calculated polydispersity indices (PDIs) and cumulant (or z-average) sizes. For example, the PDIs for the parent CNTO607 and MEDI-1912 were 0.22 and 0.15, respectively, while the PDIs for the daughter molecules were 0.1 and 0.12, respectively, indicating a more uniform monodisperse state (Table 2). Similarly, the mean particle size of the parent MEDI-1912 was 22 nm, while the mean particle size of the daughter molecule MEDI-1912-STT was 13 nm, indicating a lower-order aggregated state (Table 2). Therefore, significant self-interactions are occurring, resulting in detectable self-interactions at lower concentrations and precipitation at higher concent...

Claims

1. 1. A method for distinguishing or ranking binding agents according to their solubility in solution and / or resistance to self-association, wherein said binding agent is a transmembrane domain-containing polypeptide; (i) providing a library of higher eukaryotic clones each containing DNA encoding a binding agent; (ii) culturing the clone in vitro under conditions for expression of the binding agent, wherein the binding agent is displayed on the cell surface; (iii) determining the surface display level of said binding agent on said clones; (iv) selecting one or more clones that exhibit higher surface display of the binder compared to other clones; (v) identifying binding agents encoded by the one or more selected clones as having good solubility in solution and / or resistance to self-association; A method comprising:

2. 10. The method of claim 1, further comprising providing the selected clones for use in one or more further screening steps.

3. 3. The method of claim 1 or 2, wherein determining the surface display level of the binding agent on the clone in step (iii) is carried out by labeling the binding agent with an agent incorporating a detectable label.

4. The method of claim 3 , wherein the label is a fluorescent label.

5. The method of claim 3 or 4, wherein the agent binds to a constant region of the binding agent.

6. The method of claim 5 , wherein the binding agent comprises an Fc region and the agent binds to the Fc region.

7. 7. The method of any of claims 1 to 6, comprising sorting cells into a collect fraction and a waste fraction according to the level of surface display of a binding agent on the cells, whereby cells exhibiting surface display above a predetermined threshold are sorted into the collect fraction and cells exhibiting surface display below the predetermined threshold are sorted into a waste fraction, thereby enriching for binding agents that exhibit higher solubility in solution and / or higher resistance to self-association.

8. 8. The method of claim 7, wherein the waste fraction contains cells expressing a comparison polypeptide having a critical concentration of at least 10 mg / ml, and the collection fraction contains cells expressing a binding agent having a critical concentration at least 1.5 times higher than the comparison polypeptide in the waste fraction.

9. The method of any one of claims 1 to 8, wherein step (ii) comprises culturing the clones of the library as a mixture in one vessel.

10. The method of any one of claims 1 to 8, wherein step (ii) comprises culturing each clone of the library in a separate vessel.

11. The method of any one of claims 1 to 10, wherein the binding agent is a sequence variant of the parent binding agent.

12. 12. The method of claim 11, wherein the parent binding agent is identified as needing improved solubility in solution or resistance to self-association.

13. 13. The method of claim 11 or claim 12, wherein the method comprises generating sequence variants of the parent binding agent and integrating DNA encoding the sequence variants into cellular DNA of higher eukaryotic cells to provide a library of cellular clones comprising DNA encoding the binding agent.

14. 14. The method of claim 13, comprising: analyzing the parent polypeptide sequence; identifying one or more amino acid residues predicted to promote self-association and / or reduce solubility; and generating mutations at the one or more amino acid residues.

15. the parent binding agent has a critical concentration in phosphate buffered saline solution (PBS) of less than 50 mg / ml and / or a solubility limit in phosphate buffered saline solution (PBS) of less than 50 mg / ml; and / or the method comprises identifying a binding agent encoded by the one or more selected clones as having a critical concentration and / or solubility limit at least 1.5 times higher than that of the parent binding agent.

16. 16. The method of any one of claims 1 to 15, comprising predicting the hydrophilicity of binders based on their surface display level on said cell clones and / or identifying binders of one or more selected clones as being more hydrophilic.

17. The method of any one of claims 1 to 16, wherein expression of the DNA encoding the binding agent is under the control of a strong promoter.

18. determining the sequence of the DNA encoding the binding agent from the one or more selected clones; providing an isolated nucleic acid encoding said binding agent.

19. determining the sequence of the DNA encoding the binding agent from the one or more selected clones; 19. The method of any one of claims 1 to 18, further comprising expressing DNA encoding the binding agent in a host cell in vitro under conditions for secretion of the binding agent in soluble form.

20. 20. The method of claim 19, comprising formulating the binding agent in a composition comprising a pharmaceutically acceptable excipient.

Images