Modified immunoglobulin Fc regions

Abstract

To provide proteins containing a variant immunoglobulin Fc region that have mutations that essentially eliminate binding to Fc receptors.SOLUTION: Provided is a protein comprising a variant IgG Fc region, the protein exhibiting significantly reduced binding to human FcγRI compared to a reference protein comprising amino acid substitutions L234A / L235A / P329G. Also provided are compositions, treatment methods, and methods to reduce Fc-induced effector functions in a parent protein.SELECTED DRAWING: None

Description

[Technical Field]

[0001] The present invention relates to proteins comprising variant immunoglobulin Fc regions with mutations that essentially abolish binding to Fc receptors. [Background technology]

[0002] Antibodies are typically composed of structurally and functionally distinct regions: the variable regions are responsible for antigen binding, whereas the constant regions interact with various ligands such as C1q and Fc receptors and confer specific physiological properties or effector functions to the antibody.

[0003] In humans, there are three classes of Fcγ receptors that interact with IgG to initiate immune responses: FcγRI (CD64), FcγRII (CD32), and FcγRIII (CD16) (Bruhns 2009). One or more of these receptors are found on the surface of almost all leukocytes, except T cells. Responses they mediate include cytokine release, B cell activation and differentiation, endocytosis, phagocytosis, and cytotoxicity. Each receptor is characterized by an alpha chain that binds to the IgG Fc region and associates with signaling motifs either in the cytoplasmic portion of the alpha chain or on associated membrane proteins.

[0004] FcγRI binds human IgG1 and IgG3 with relatively high affinity and IgG4 with lower affinity. It is found on monocytes and macrophages and is involved in the phagocytosis of immune complexes and the release of immune mediators. FcγRII exists in three forms with distinct functions and cellular distributions. They have low affinity for IgG1 and IgG3 but bind aggregated IgG more avidly. FcγRIIA is expressed on monocytes, granulocytes, and platelets. It contains ITAM motifs in its cytoplasmic tail that mediate cell activation. One allele also binds IgG2. FcγRIIB is found on B cells and contains ITIM motifs that block cell activation. There are two forms of FcγRIII. Both have intermediate affinity for monomeric IgG1 and IgG3. FcγRIIIA, like other FcRs, is a transmembrane glycoprotein. It is found on monocytes, macrophages, NK cells, and some T cells, where it associates with a separate membrane protein containing ITAM motifs. FcγRIIIB is a lipid-anchored glycoprotein found on neutrophils. There are different alleles of both forms of FcγRIII that differ in IgG binding and / or functional activity. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] WO199958572 [Patent Document 2] WO2006076594 [Patent Document 3] WO2006047350 [Patent Document 4] WO2006053301 [Patent Document 5] WO2011066501 [Patent Document 6] WO2013165690 [Patent Document 7] WO2012130831 [Patent Document 8] WO2014108483 [Patent Document 9] US20060235208 [Patent Document 10] US6194551 [Patent Document 11] US8969526 [Patent Document 12] US10011660 [Patent Document 13] WO2000042072 [Patent Document 14] WO2006033386 [Patent Document 15] WO2008145142 [Patent Document 16] WO2010063785 [Patent Document 17] US2017029521 [Patent Document 18] WO2012022982 [Non-patent literature]

[0006] [Non-Patent Document 1] Dictionary of Microbiology and Molecular Biology, 3rd edition, 2006, John Wiley & Sons Ltd [Non-patent document 2] Oxford Dictionary of Biochemistry and Molecular Biology, 2nd Edition, 2006, Oxford University Press Summary of the Invention [Problem to be solved by the invention]

[0007] Surprisingly, the inventors have identified proteins containing variant Fc regions comprising a new set of amino acid substitutions that result in significantly lower binding to FcγRI than that seen in known variant Fc regions with discernible FcγRI binding. The inventors provide proteins comprising a variant IgG Fc domain or region, which exhibit significantly reduced binding to human FcγRI compared to a reference protein comprising the amino acid substitutions L234A / L235A / P329G. Compositions, methods of treatment, and methods for reducing Fc-induced effector function in a parent protein are also provided. [Means for solving the problem]

[0008] In a first aspect of the present invention, there is provided a protein comprising a variant Fc region or a variant Cγ2 domain, wherein the variant Fc region or variant Cγ2 domain comprises: (a) an amino acid substitution at position 234, or an amino acid substitution at position 235, or an amino acid substitution at both positions 234 and 235; and (b) an amino acid change at position 236 to arginine (R), wherein amino acid numbering is according to the EU index as set forth in Kabat, and wherein binding of the protein to human FcγRI is significantly reduced compared to binding of a LALAPG reference protein.

[0009] In a second aspect of the present invention there is provided a protein comprising a variant IgG Fc region, wherein the variant IgG Fc region comprises a combination of amino acid substitutions selected from the group defined in claim 19, wherein the amino acid numbering is according to the EU index as set out in Kabat.

[0010] In a third aspect of the present invention, there is provided a nucleic acid comprising a sequence encoding a protein as defined above.

[0011] In a further aspect, there is provided a composition, vector or host cell comprising a nucleic acid as defined above.

[0012] In a further aspect, there is provided a method of producing a protein as defined above, comprising the steps of: (a) culturing a host cell as defined above; and (b) isolating the protein.

[0013] In a further aspect, there is provided a conjugate or composition comprising a protein as defined above.

[0014] In a further aspect, there is provided a method for reducing Fc-mediated effector function of a protein comprising an IgG Fc region, the method comprising the steps of: (a) engineering the synthesis and expression of a gene encoding a protein as defined above; and (b) isolating the protein.

[0015] In a further aspect, there is provided a method of treating a mammal, comprising administering to the mammal an effective amount of a protein, conjugate, composition or vector as defined above.

[0016] In a further aspect, there is provided a protein, conjugate, composition or vector as defined above for use as a medicament, wherein the reduced effector function of said protein (a) reduces the likelihood of an adverse clinical effect, (b) increases the likelihood of a beneficial clinical effect, or (c) facilitates the use of higher, more frequent or greater doses compared to a reference protein.

[0017] In a further aspect, there is provided a protein, conjugate, composition or vector as defined above for use in a test system for research or diagnostics or quality control, such as for use in an enzyme-linked immunosorbent assay (ELISA), a fluorescence-based assay, an immunohistochemistry test or an imaging test.

[0018] The invention will now be described with reference to the following non-limiting drawings. [Brief explanation of the drawings]

[0019] [Figure 1]

[0023] Figure 1 shows an alignment of the amino acid sequences of wild-type immunoglobulin Fc regions from human IgG2, human IgG2, human IgG4, mouse IgG2a, rat IgG2b, and rabbit IgG. Residues are numbered according to the EU index. Amino acid residues that are altered in the various variants described herein are highlighted in boxes. [Figure 2] Figure 1 shows the dose response to human IgG in the NanoBiT® FcRn immunoassay. Luminescence response (average of duplicates) is plotted against the concentration of human IgG. [Figure 3] Figure 1 shows dose response to wild-type CD20 reference antibody in an ADCC assay. Luminescence responses (average of 8 replicates) are plotted as a percentage of the response at 1000 ng / mL, along with the standard deviation of the mean. [Figure 4] Figure 1 shows dose response to wild-type CD20 reference antibody in an ADCP assay. Luminescence responses (average of four replicates) are plotted as a percentage of the response at 1000 ng / mL, along with the standard deviation of the mean. [Figure 5] 5A-5D show cytokine release from peripheral blood mononuclear cells from five healthy donors after stimulation with wild-type or variant CD3 antibodies at a final concentration of 10 μg / mL. [Figure 6] Figure 6 shows the analysis of carbohydrates released from samples of wild-type and variant CD20 antibodies by hydrophilic interaction liquid chromatography. Samples as described in Table 5 were prepared from HEK cells (Figure 6A) or HEK and CHO cells (Figure 6B). DETAILED DESCRIPTION OF THE INVENTION

[0020] definition The meaning of certain terms used in this invention may differ from the meaning of similar terms used in the literature. Definitions of such terms are provided below in the context of this invention.

[0021] Unless otherwise specified, all technical and scientific terms have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. For example, Dictionary of Microbiology and Molecular Biology, 3rd Edition, 2006, John Wiley & Sons Ltd and Oxford Dictionary of Biochemistry and Molecular Biology, 2nd Edition, 2006, Oxford University Press provide general references to many of the terms used in this invention.

[0022] Amino acids can be referred to by either their commonly known three letter symbols or by the one-letter symbols recommended by the IUPAC-IUB Biochemical Nomenclature Commission. Similarly, nucleotides can be referred to by their commonly accepted one-letter symbols.

[0023] Throughout this specification and claims, the numbering of amino acid residues in immunoglobulins is that of the human IgG1 EU myeloma protein (the "EU index").

[0024] Approximately. The term "about" or "approximately" in reference to a numerical value X means, for example, X plus or minus 20%.

[0025] Affinity. The term "affinity" refers to the tendency for two molecules (e.g., an immunoglobulin Fc region and an Fc receptor) to associate (bind) non-covalently, i.e., by means of interactions that do not involve the formation of covalent chemical bonds. When two molecules have affinity for each other, it means that there is a favorable tendency for them to associate specifically non-covalently to form a complex or complexes. For purposes of this application, the magnitude of affinity can be defined relative to the ratio of the concentrations of the free molecules and the complex to obtain an association constant for the interaction under specific conditions of temperature, salt concentration, and pH. Alternatively, it can be defined as relative affinity, which is the ratio of measurements that correlate to the association constant for a molecule of interest (e.g., a variant Fc region) binding to a target molecule (e.g., an Fc receptor) and a reference molecule (e.g., a wild-type reference, a LALAPG reference, or a LALA reference) binding to the same target molecule under the same experimental conditions.

[0026] Amino acid modification. The term "amino acid modification" refers to a change in the amino acid sequence of a predetermined amino acid sequence. Exemplary modifications include substitution, insertion, or deletion of an amino acid residue. The substituted or inserted amino acid residue is a naturally occurring amino acid residue (i.e., encoded by the genetic code).

[0027] Amino acid deletion. The term "amino acid deletion" refers to the deletion of at least one amino acid in a predetermined amino acid sequence. Throughout this specification and claims, an amino acid deletion may be described as XnnnΔ, where X is the single-letter code of the residue in the predetermined amino acid sequence, nnn is the position in that sequence as defined by the EU index, and Δ (Greek letter delta) indicates that the specified amino acid residue is deleted.

[0028] Amino acid insertion. The term "amino acid insertion" refers to the insertion of at least one amino acid in a predetermined amino acid sequence. Throughout this specification and claims, an amino acid insertion may be written as ΔnnnX, where Δ (Greek letter delta) indicates that the specified amino acid residue is not present in the predetermined amino acid sequence, nnn is its position in that sequence as defined by the EU index, and X is the one-letter code of the residue being inserted.

[0029] Amino acid substitution. The term "amino acid substitution" refers to the replacement of at least one amino acid residue in a predetermined amino acid sequence. The replacement residue is a naturally occurring amino acid residue (i.e., encoded by the genetic code). Throughout this specification and claims, amino acid substitutions may be described as XnnnY, where X is the one-letter code of the residue in the predetermined amino acid sequence, nnn is the position in that sequence as defined by the EU index, and Y is the one-letter code of the replacement amino acid residue.

[0030] Antibodies. The term "antibody" includes immunoglobulin molecules or fragments thereof having one or more protein chains containing at least one domain capable of binding to an antigen (an "antigen-binding domain"). Immunoglobulins can be of any isotype (e.g., IgG, IgE, IgM, IgD, IgA, or IgY), subtype (e.g., IgG1, IgG2, IgG2a, IgG2b, IgG2c, IgG3, IgG4, IgA1, and IgA2), or allotype, or they can be hybrids from more than one isotype, subtype, or allotype. Immunoglobulins can be of any species (e.g., human, monkey, camel, llama, goat, sheep, rabbit, mouse, rat, mouse, hamster, or chicken), or they can be hybrids from more than one species. Antibodies can be polyclonal or monoclonal. Antibodies can be naturally occurring or non-naturally occurring (i.e., isolated antibodies). The antibody may be genetically engineered (e.g., chimeric, humanized, camelized, intracellular, or bispecific). The antibody may also be a fragment (e.g., a Fab fragment, a F(ab')2 fragment, a Fv fragment, or a single-chain Fv fragment). According to this definition, "antibody" includes both polyclonal and monoclonal antibodies, or any mixture thereof. The protein of the invention may be an antibody or antibody fragment according to any of the definitions of "antibody" provided herein.

[0031] Antibody-Dependent Cell-Mediated Cytotoxicity. Antibody-dependent cell-mediated cytotoxicity (ADCC), also called antibody-dependent cellular cytotoxicity, is a mechanism of immune defense in which effector cells of the immune system actively lyse target cells to which an antibody or any other protein containing an Fc region binds. ADCC is mediated through the binding of the immunoglobulin Fc region to an Fc receptor on the effector cell. This can involve different types of Fc receptors and different types of effector cells, but is commonly mediated through FcγRIIIA expressed on natural killer (NK) cells.

[0032] Antibody-dependent cell-mediated phagocytosis. Antibody-dependent cell-mediated phagocytosis (ADCP) is a mechanism of immune defense in which effector cells of the immune system phagocytose target cells bound by antibodies or any other proteins containing an Fc region. ADCP can be mediated by monocytes, macrophages, neutrophils, or dendritic cells via FcγRIIA (CD32A), FcγRI (CD64), and / or FcγRIIIA (CD16A).

[0033] Binding. "Binding," in the context of this invention, refers to a reversible interaction, typically between macromolecules such as proteins. In the simplest case, the binding interaction between two macromolecules A and B can be represented by the following equation:

[0034]

number

[0035] According to the law of mass action, the rate of the forward reaction is given by the rate constant k a multiplied by the concentrations of the separate reactants, i.e., k a × [A] × [B], and the rate of the reverse reaction is given by the rate constant k d multiplied by the concentration of the complex, i.e., k d × [AB] (where the square brackets indicate the concentrations of the reactants). At equilibrium, the rates of both reactants are the same and the binding affinity is given by the association constant K a As:

[0036]

number

[0037] or dissociation constant K d As:

[0038]

number

[0039] The rate constants of the two reactions can be defined as the ratio of the rate constants of the two reactions.

[0040] K d is simply K a More commonly in the literature, K d is used as a measure of affinity. The strength of the binding interaction is measured by the association constant K a or dissociation constant K d Alternatively, and particularly for comparing binding of different substances as in the examples of this invention, binding can be determined by measuring the response to a concentration of complex (AB) at fixed concentrations of reactants A and B. For example, binding may be measured by surface plasmon resonance (SPR) as the increase in SPR response after injecting a fixed concentration of one reactant (e.g., an antibody) onto a surface to which a fixed concentration of the other reactant (e.g., an Fc receptor) is attached. Or again, binding may be measured by enzyme-linked immunosorbent assay (ELISA) as the absorbance provided by enzymatic conversion of a substrate (e.g., TMB) after a series of steps that may include, for example, (a) adsorbing a reactant (e.g., an antibody) to a microplate, (b) blocking nonspecific binding sites, and (c) incubating with a second reactant (e.g., C1q) labeled with an enzyme (e.g., horseradish peroxidase).

[0041] Relative binding can be a ratio of measured binding of a molecule of interest (e.g., a variant Fc region) that binds to a target molecule (e.g., an Fc receptor) to a reference molecule (e.g., a wild-type reference, a LALAPG reference, a LALA reference) or buffer alone that binds to the same target molecule under the same experimental conditions.

[0042] Charged Amino Acid Residues. A "charged amino acid residue," in the context of this invention, is one of aspartic acid (D), glutamic acid (E), histidine (H), lysine (K), or arginine (R). A negatively charged amino acid residue is either aspartic acid (D) or glutamic acid (E). A positively charged amino acid residue is either histidine (H), lysine (K), or arginine (R). The "net change in charged amino acid residues" as a result of an amino acid substitution or modification is the absolute value of the difference between the net number of charged residues contained in a protein comprising a variant Fc region of the invention and the net number of charged residues contained in its corresponding wild-type reference protein, where the net number of charged residues is the number of all negatively charged residues minus the number of all positively charged residues.

[0043] Complement System. In the context of this invention, the "complement system," also known in other contexts as "complement," refers to a part of the innate immune system that includes numerous plasma proteins that can be recruited and activated by antibodies (or other proteins containing Fc regions) to promote inflammation and attack the plasma membrane of target cells. The classical complement pathway begins with activation of the first component, the C1 complex. The C1 complex contains three types of polypeptide chains: C1q, C1r, and C1s. The Fc region of an IgM or IgG antibody binds to C1q, initiating the activation of C1r and C1s. The subsequent cascade of reactions can result in the deposition of the membrane attack complex on the surface of target cells.

[0044] Complement-dependent cytotoxicity. Complement-dependent cytotoxicity (CDC), also called complement-mediated cytotoxicity (CMC), is a mechanism of immune defense in which proteins containing an Fc region (e.g., antibodies) bound to a target cell initiate activation of the complement system, resulting in lysis of the target cell. CDC is mediated through the binding of the Fc region to C1q, resulting in activation of C1r and C1s, and a subsequent cascade of reactions that can lead to the deposition of the membrane attack complex on the surface of the target cell.

[0045] Composition. A "composition," in the context of this invention, is a mixture containing both an active substance intended to have a pharmacological or physiological effect and an inactive substance not intended to have a pharmacological or physiological effect. The inactive substance may be useful for providing a stable formulation of the active substance. The composition may be, for example, a pharmaceutical product containing a drug substance.

[0046] The term "comprise" means "consisting of" in addition to "include." For example, a composition "comprising" X and Y may consist exclusively of X and Y, or it may include something additional, such as X+Y+Z.

[0047] Conjugate (noun). A "conjugate" or "conjugated protein," in the context of this invention, is a protein to which another substance has been covalently attached. The other substance may itself comprise a protein, or it may comprise any other type of molecule or macromolecule. The other substance may be useful for a variety of purposes, for example, useful for treating a disease, useful for diagnosing a disease, or useful for purification or analysis.

[0048] Consensus Method. The term "consensus method" in the context of this invention refers to the IEDB version 2.22 consensus method for determining peptide binding to MHC class II alleles as described by Wang 2008.

[0049] Constant Region. The term "constant region" refers to the immunoglobulin domains excluding the variable domains. For example, the constant region may comprise a single constant domain of a light chain, or three constant domains of an IgA, IgD, or IgG heavy chain, or four constant domains of an IgE or IgM heavy chain. The heavy chain constant region may comprise a flexible hinge region between the first and second constant domains. The IgA or IgM heavy chain constant region may comprise the J chain.

[0050] Culture Medium. A "culture medium" is an aqueous mixture capable of supporting the growth of cells.

[0051] Cytokines. A "cytokine" is a member of a large group of proteins secreted by cells and that play a role in signaling between different cells. Many of them bind to receptors to modulate, for example, humoral or cellular immune responses, and are involved in the growth and differentiation of numerous cell types in various ways. Cytokines include, for example, chemokines, interferons, interleukins, lymphokines, and tumor necrosis factors.

[0052] Cytokine release. The term "cytokine release," in the context of this invention, refers to the release of proinflammatory cytokines from cells of the immune system exposed to a protein comprising an Fc region. Proinflammatory cytokines include, for example, one or more of GM-CSF, TNFα, IFNγ, IL-2, IL-4, IL-6, IL-8, and IL-10.

[0053] Domain. A protein "domain" is a conserved portion of a given protein sequence and tertiary structure that can develop, function, and exist independently of the rest of the protein chain. Each domain forms a compact three-dimensional structure that is often independently stable and can fold. Many proteins consist of several structural domains. Antibodies generally have several immunoglobulin domains in each polypeptide chain. Variable domains have a high degree of sequence variability and are responsible for binding to antigens. Constant domains are more highly conserved and are responsible for the structural and functional integrity of antibodies.

[0054] Effector function. "Effector function," in the context of this invention, refers to any one or more physiological effects that can be mediated by the Fc region in vitro or in vivo. Such effects include, for example, binding to C1q, activation of complement, complement-dependent cytotoxicity, binding to FcγRI, binding to FcγRII, binding to FcγRIII, antibody-dependent cell-mediated cytotoxicity, antibody-dependent cell-mediated phagocytosis, or induction of cytokine release.

[0055] EU Index. "EU index" or "EU index as described in Kabat" or "EU numbering scheme" refers to the numbering of EU myeloma proteins (Edelman 1969; Kabat 1991).

[0056] Fc region. The term "Fc region" refers to the constant region of an immunoglobulin heavy chain excluding the first constant domain. For example, the Fc region may include the last two constant domains of an IgA, IgD, or IgG heavy chain, or the last three constant domains of an IgE or IgM. The Fc region may include a flexible hinge at the N-terminus of these domains. An IgA or IgM Fc region may also include a J chain. Thus, an IgG Fc region typically includes constant domains Cγ2 and Cγ3, and may include a hinge between Cγ1 and Cγ2. Although the boundaries of the Fc region are variable, a human IgG Fc region with a hinge typically includes residue E216 at its carboxyl terminus. The human Cγ2 domain typically includes residues A231 to K340. The human IgG Cγ3 domain typically includes residue G341 at the carboxyl terminus. The Fc region may be monomeric, but is more commonly dimeric or multimeric.

[0057] Fc variants. The term "Fc variant" refers to a protein comprising a variant Fc region. This may be an immunoglobulin comprising a variant Fc region, or a fusion protein in which one of the components of the protein comprises a variant Fc region.

[0058] Fc Receptors. An "Fc receptor" (FcR) is a protein found on the surface of certain types of cells, such as B lymphocytes, macrophages, natural killer cells, dendritic cells, neutrophils, basophils, eosinophils, and platelets, that binds to the Fc region of immunoglobulins. There are several types of Fc receptors, which are classified according to the class of immunoglobulin they bind. For example, Fc-gamma receptors (FcγR) bind IgG, Fc-epsilon receptors (FcεR) bind IgE, and Fc-alpha receptors (FcαR) bind IgA. In humans, there are various subtypes of FcγR, e.g., FcγRI (CD64), FcγRIIA (CD32A), FcγRIIB (CD32B), FcγRIIC (CD32C), FcγRIIIA (CD16A), and FcγRIIIB (CD16B), which differ in their affinity and specificity for IgG subclasses. Furthermore, there are various allelic forms and other variants. In addition to FcγR, the neonatal Fc receptor (FcRn or Brambell receptor) is expressed on multiple cell types and is involved in IgG transport, e.g., across the placenta or to the mammary gland, as well as in the regulation of IgG catabolism and homeostasis of IgG serum levels. Similar Fc receptors have been found in other mammals, although the number and distribution of types, subtypes, and alleles may vary among different species.

[0059] Fusion Proteins. A "fusion protein" (sometimes known as a chimeric protein) is a protein created by combining two or more genes that originally coded for separate proteins. Transcription and translation of this fusion gene results in a single or multiple proteins with properties derived from each of the original proteins. Naturally occurring fusion proteins can be found in cancer cells. The term "fusion protein," in the context of this invention, is intended to mean a fusion protein created by genetic engineering, which may contain genes or gene segments that have been altered (compared to the original genes that coded for the separate proteins). A fusion protein may contain a single or multiple proteins, any of which may be post-translationally modified.

[0060] Glycoform. "Glycoform" refers to any one of a number of different forms of a glycoprotein that differ with respect to the number, type, or structure of the attached carbohydrates. Immunoglobulin Fc regions typically have a carbohydrate attached to Asn297. The various structures of this carbohydrate give rise to different glycoforms of a protein comprising an Fc region. In the context of this invention, a "glycoform spectrum" can be determined after release of the carbohydrate from a protein comprising an Fc region, followed by separation of the carbohydrate mixture by suitable chromatographic techniques that resolve the different structures and allow comparison of different samples by graphical or mathematical methods.

[0061] Heavy chain. The term "heavy chain" refers to a high molecular weight protein subunit of an immunoglobulin. The heavy chain may be of any immunoglobulin isotype (e.g., IgG, IgE, IgM, IgD, IgA, or IgY), subtype (e.g., IgG1, IgG2, IgG2a, IgG2b, IgG2c, IgG3, IgG4, IgA1, or IgA2), or allotype.

[0062] Host Cell: A "host cell" is a cell that can be transfected with a nucleic acid to produce a cell that expresses the product of a gene contained within the nucleic acid.

[0063] Immune Epitope Database. The "Immune Epitope Database" (IEDB) is a resource provided by the National Institute of Allergy and Infectious Diseases and is available at www.iedb.org (Dhanda 2019). It lists experimental data on antibody and T cell epitopes studied in humans and other animal species in the context of infectious disease, allergy, autoimmunity, and transplantation. The IEDB also curates tools to aid in the prediction and analysis of B-cell and T-cell epitopes, including peptides with MHC-binding properties.

[0064] Immunoglobulin. In the context of this invention, "immunoglobulin" describes a molecule having one or more protein chains containing at least one domain capable of binding to an antigen (an "antigen-binding domain") and at least one Fc region. Immunoglobulins may be of any isotype (e.g., IgG, IgE, IgM, IgD, IgA, or IgY), subtype (e.g., IgG1, IgG2, IgG2a, IgG2b, IgG2c, IgG3, IgG4, IgA1, and IgA2), or allotype. Immunoglobulins may be of any species (e.g., human, monkey, camel, llama, goat, sheep, rabbit, mouse, rat, mouse, hamster, or chicken) or may be hybrids derived from more than one species. Immunoglobulins may be polyclonal or monoclonal. Immunoglobulins may be naturally occurring or genetically engineered (eg, chimeric antibodies, humanized antibodies, camelized antibodies, intracellular antibodies).

[0065] Label: The term "label" refers to any organic or inorganic molecule or molecular complex that can be covalently attached to a protein and that can also be detected by virtue of its intrinsic properties, such as color, fluorescence, radioactivity, luminescence, catalytic activity, size, or mass.

[0066] Light chain. The term "light chain" refers to a low molecular weight protein subunit of an immunoglobulin. The light chain may be of any type (e.g., kappa or lambda), subtype, or allotype of light chain.

[0067] Major Histocompatibility Complex. The "major histocompatibility complex" (MHC) is a set of genes encoding cell surface proteins essential for the adaptive immune system of vertebrates to recognize foreign molecules. The MHC gene family is divided into three subgroups: MHC class I, MHC class II, and MHC class III. MHC class I molecules can be recognized by the CD8 coreceptor, while MHC class II molecules can be recognized by the CD4 coreceptor. Both contain (a) an alpha chain that binds antigen-derived peptides, and (b) a beta2 microglobulin chain. The combination of a peptide and a specific MHC molecule can be recognized by a T cell receptor and may initiate the immune function of the T cell. In humans, MHC molecules are also known as human leukocyte antigens (HLA). In the context of this invention, reference to "MHC" will be understood as a reference to "human MHC" or "HLA" unless otherwise specified.

[0068] Nucleic Acids. Nucleic acids are biological macromolecules containing linear chains of nucleotides linked by phosphodiester bonds. Nucleic acids include both deoxyribonucleic acid (DNA) and ribonucleic acid (RNA).

[0069] Pharmacokinetic Parameters. Pharmacokinetics (PK) is the branch of pharmacology dedicated to determining the fate of substances administered to an organism. It describes and measures the processes of absorption, distribution, and elimination (i.e., metabolism and excretion). These processes can be mathematically modeled in a variety of ways, for example, compartmental and non-compartmental models. Various "pharmacokinetic parameters" (otherwise known as pharmacokinetic metrics) can be determined from the analysis of concentration (C) as a function of time (t) after administration of a specific amount of drug (D). These include, for example, the following:

[0070] [Table 1]

[0071] A number of pharmacokinetic parameters and methods for determining them are described in Rowland 1995.

[0072] Protein. A "protein" is a macromolecule comprising one or more chains of amino acid residues linked together by peptide bonds. Proteins are typically produced by ribosomal translation of messenger RNA. Proteins may additionally contain various post-translational modifications. These include, for example, lipidation, myristoylation, palmitoylation, isoprenylation, prenylation, farnesylation, geranylgeranylation, glypiation, lipoylation, flavin attachment, heme attachment, phosphopantetheinylation, Schiff base formation, acylation, acetylation, alkylation, amidation, amide bond formation, butyrylation, glycosylation, malonylation, hydroxylation, iodination, nucleotide addition, phosphorylation, adenylylation, uridinylylation, proprionylation, pyroglutamate formation, S-glutathionylation, S-nitrosylation, S-sulfenylation, S-sulfinylation, S-sulfonylation, succinylation, sulfation, glycation, carbonylation, isopeptide bond formation, biotinylation, carbamylation, disulfide bond formation, and PEGylation.

[0073] Reference Protein. A "reference protein," in the context of this invention, is a protein comprising an Fc region to which a protein of the invention can be compared with respect to one or more functional properties, such as, for example, binding to an Fc receptor, binding to C1q, thermostability, aggregation, glycosylation, immunogenicity, protease susceptibility, etc. For example, the reference protein can be (a) a protein comprising a human IgG1 Fc region having the amino acid sequence specified in SEQ ID NO: 1 ("wild-type reference" or "wild-type reference protein"), (b) a protein comprising a human IgG1 Fc region having the amino acid sequence specified in SEQ ID NO: 2 ("LALAPG reference" or "LALAPG reference protein"), or (c) a protein comprising a human IgG1 Fc region having the amino acid sequence specified in SEQ ID NO: 3 ("LALA reference" or "LALA reference protein").

[0074] Significantly reduced. "Significantly reduced," in the context of this invention, means reduced to the extent that the probability of occurring by chance alone is less than 5%, as assessed by an appropriate statistical test.

[0075] Stability: "Stability" in the context of this invention refers to the absence of change in a measured property over time.

[0076] Substantially reduced. "Substantially reduced" in the context of this invention means reduced to a significant degree, for example reduced by 50%, 60%, 70%, 80%, 90%, 95%, 98%, 99% or 100%.

[0077] Surface Plasmon Resonance. "Surface Plasmon Resonance" (SPR) is the resonant oscillation of conduction electrons at the boundary between negative and positive dielectric constant materials stimulated by incident light. In the context of this invention, the term refers to a technique based on the SPR principle for measuring non-covalent interactions between macromolecules such as proteins. There are many implementations that incorporate this technology, such as Biacore®.

[0078] Thermal Stability. The term "thermostability" refers to the ability of a substance, such as a protein, to resist irreversible changes in its chemical or physical structure at elevated relative temperatures. The thermal stability of a protein can be measured by determining the temperature at which an irreversible change, such as unfolding, denaturation, aggregation, or precipitation, occurs. Alternatively, it can be measured by determining the rate at which such a change occurs at a particular temperature.

[0079] Variable region. The term "variable region" refers to one or more variable domains of an immunoglobulin heavy or light chain. Although the boundaries of the variable domain can vary, human heavy-chain variable domains usually include residue 117 from the amino terminus, and human light-chain variable domains include residues R108 (kappa light chains) or G107a (lambda light chains) from the amino terminus.

[0080] Variant Fc Region. The term "variant Fc region" refers to either (a) an Fc region with the same amino acid sequence as a naturally occurring Fc region, except for an amino acid alteration at one or more positions up to 12, or (b) an Fc region with the same amino acid sequence as any previously disclosed Fc region, except for an amino acid alteration at one or more of positions 234, 235, and 236 according to EU numbering.

[0081] Variant Cγ2 domain. The term "variant Cγ2 domain" refers to (a) a Cγ2 domain that has the same amino acid sequence as a naturally occurring Cγ2 domain, except for an amino acid change at one or more positions, up to a maximum of 12, or (b) a Cγ2 domain that has the same amino acid sequence as any previously disclosed Cγ2 domain, except for an amino acid change at one or more of positions 234, 235, and 236 according to EU numbering.

[0082] Vector. In the context of this invention, a "vector" is a polynucleotide used as a vehicle to artificially deliver foreign genetic material into cells, where it can be replicated and / or expressed. The cells may be prokaryotic, but are preferably eukaryotic. The vector may be a plasmid, viral vector, cosmid, or artificial chromosome. Vectors for expressing proteins in eukaryotic host cells typically contain (a) sequences necessary for replication in bacteria, including an origin of replication and an antibiotic resistance gene (e.g., for resistance to kanamycin or ampicillin), (b) sequences necessary for expression of the foreign genetic material in eukaryotic cells, including a multiple cloning site for insertion of the foreign genetic material, at least one promoter sequence for driving expression of the foreign genetic material, a polyadenylation signal, a ribosome recognition site, e.g., a Kozak sequence, and, optionally, (c) a sequence encoding a selectable marker in eukaryotic cells, such as resistance to antibiotics, such as G418, hygromycin, or puromycin.

[0083] Adaptability to various IgG Fc regions and domains Throughout this specification, for simplicity and clarity, reference is frequently made to variant Fc regions of human IgG1 antibodies containing one or more amino acid changes at various positions (according to the EU numbering system). This is not intended to limit the scope of the present invention in any way. Amino acid residues at equivalent positions in other subclasses of human IgG, including IgG2, IgG3, and IgG4, or in any subclass of IgG from other species, can be similarly altered to produce the desired effect. Other species include, for example, birds and mammals, such as primates, rodents, lagomorphs, carnivores, and artiodactyls. Also included within the scope of this specification are naturally occurring allotypes of IgG, as well as variants containing other mutations, such as those that may be introduced to alter binding to human FcRn or to improve other structural or functional properties, such as stability or elimination of unwanted heterogeneity. Furthermore, this specification is not limited to variant Fc regions of intact antibodies, but also includes variant Fc regions contained within any other protein, such as fusion proteins, immunoadhesins, immunocytokines, single-chain antibodies, bispecific antibodies, multispecific antibodies, bispecific T cell engagers, antibody-drug conjugates, and enzyme replacement-Fc fusion proteins. In addition, this specification is not limited to intact or complete variant Fc regions, but also includes variants that include any portion of an Fc region that is predicted to be capable of binding to an Fcγ receptor in its wild-type form, such as the Cγ2 domain, or a combination of the hinge region and the Cγ2 domain. Thus, wherever the context allows, the term "Fc region" should be understood to include not only intact Fc regions, but also any portion of an Fc region that is predicted to be capable of binding to an Fcγ receptor in its wild-type form.

[0084] Binding of the Fc region to Fcγ receptors In many situations, it is desirable to reduce, or preferably eliminate, IgG binding to Fcγ receptors because the biological effects they mediate are unnecessary and potentially harmful (see, e.g., Wang 2018). For example, when an antibody is used to neutralize the activity of an antigen, the concomitant activation of a cellular response may be undesirable. However, the Fc region has other useful properties, notably its ability to provide a long half-life based on binding to FcRn. This feature is used in the development of fusion proteins, in which the Fc region confers extended half-life to other biologically active proteins. Nevertheless, activation of a cellular response may be undesirable. Therefore, researchers have developed variant forms of Fc with mutations that selectively reduce binding to FcγR.

[0085] Site-directed mutagenesis and X-ray crystallography of the Fc / FcγRIII complex have enabled the identification of amino acid residues in the Fc region involved in binding to Fc receptors. Many different residues have been identified whose mutations could result in a reduction in binding to one or more Fc receptors. However, no single mutation has been found to completely abolish FcγR binding, and researchers have therefore attempted to identify combinations of mutations (preferably as few as possible) that have the desired effect. The sheer number of possible sites and the possibility of deleting or converting them to any one of 19 alternative amino acid residues means that the number of possible mutation combinations is too large to practically test exhaustively. Nevertheless, it has been proposed that certain combinations can "completely abolish" immune effector function (e.g., Schlothauer 2016). Surprisingly, we now find that Fc variants containing combinations of mutations previously considered inactive still have detectable binding activity. Therefore, there is a need for a minimal set of mutations that truly eliminates human Fc binding to Fcγ receptors.

[0086] Amino acid changes at a single site Various investigators have used site-directed mutagenesis to explore the role of individual amino acid residues in the binding of the human IgG1 Fc region to human FcγRs. Table 1 summarizes a non-limiting selection of results.

[0087] [Table 2]

[0088] Amino acid changes at multiple sites Thus, from the foregoing, there are at least 30 positions in the IgG1 Fc region where mutations could potentially reduce binding to FcγR. However, none have been shown to completely abolish binding to all of the receptors. In an effort to identify variants that more effectively abolish binding, researchers have tested combinations of mutations at different positions. See, e.g., WO199958572, WO2006076594, WO2006047350, WO2006053301, WO2011066501, WO2013165690, WO2012130831, WO2014108483, US20060235208, US6194551, US8969526, US10011660; Xu 2000; Hezarah 2001, Shields 2001, Lin 2013; Dall'Acqua 2006, Oganesyan 2008, Schlothauer 2016, Tam 2017.

[0089] The number of possible combinations corresponding to the identified mutations (Table 1) is extremely large (approximately 5 × 10 17 ). Clearly, it would be completely impractical to generate and test all of the possible alternative substitution combinations at each site, let alone the vast number of them. In practice, even fewer combinations have been tested. Of these, few have been used in therapeutic antibodies entering clinical trials, as exemplified in Table 2.

[0090] [Table 3]

[0091] Binding of the Fc region to FcRn The neonatal Fc receptor (FcRn) is widely expressed on epithelial cells, where it is involved in mediating the transport and recycling of IgG (Kuo 2011). FcRn-mediated transport of IgG begins with fluid-phase pinocytosis of IgG at the plasma membrane, followed by binding to FcRn under the acidic conditions of the endosome. From there, IgG can be transcytosed to the opposite membrane surface (e.g., for transport across the placenta) or recycled back to the original membrane surface. As a result of FcRn activity, IgG has a longer half-life than it would otherwise have. This is a distinct advantage for the use of therapeutic agents containing the Fc region, and any modification that reduces the ability of IgG to bind to Fcγ receptors (or C1q) without adversely affecting binding to FcRn is desirable. Indeed, many researchers have attempted to enhance binding to FcRn in order to increase drug half-life (e.g., Dall'Acqua 2006; Zalevsky 2010). Because the binding site of FcRn is located in the CH2-CH3 domain and is distinct from the binding sites of Fcγ receptors and C1q, the idea is to identify different amino acid changes that affect binding of one type of receptor but not the other.

[0092] Measurement of binding to Fc receptors A variety of different techniques have been used to measure the binding of variant immunoglobulins to Fc receptors, such as FcγR and FcRn, including enzyme-linked immunosorbent assays (ELISAs) using soluble recombinant Fc receptors (e.g., WO2000042072; Shields 2001), flow cytometry using mammalian cells transfected with genes encoding Fc receptors (e.g., WO2000042072; Shields 2001), and immunohistochemistry (e.g., WO2000042072; Shields 2001). 2001), AlphaScreen® (a bead-based luminescent proximity assay) using tagged soluble recombinant Fc receptors (e.g., US20060235208; WO2006047350, WO2011066501), multiplexed microsphere (Luminex®) assays (e.g., Boesch), surface plasmon resonance using the Proteon® system with soluble recombinant Fc receptors (e.g., WO2013165690), and surface plasmon resonance using the Biacore® system with soluble recombinant Fc receptors (e.g., WO2011066501, WO2012130831; Schlothauer2016). Other experimental techniques and other implementations of the aforementioned techniques are also known in the art. The wide variety of experimental techniques, experimental conditions and different preparations of variant immunoglobulins and Fc receptors makes it difficult or impossible to reliably compare results from different studies to determine whether one particular combination of mutations is indeed more effective than another for the purpose of silencing binding to FcγRs.

[0093] Binding of the Fc region to C1q Complement component 1q (C1q) is a protein complex consisting of six copies of three protein chains each assembled into a globular head attached to a stalk. It binds to antibodies and initiates a cascade of reactions that lead to activation of the complement system.

[0094] Amino acid changes at a single site Various researchers have used site-directed mutagenesis to explore the role of individual amino acid residues in the binding of the human IgG1 Fc region to human C1q. Table 3 summarizes a non-limiting selection of results.

[0095] [Table 4]

[0096] Amino acid changes at multiple sites Therefore, from the above, there are at least 14 positions in the IgG1 Fc region that can be mutated to reduce C1q binding. However, none of them have been shown to completely abolish binding. In order to identify variants that more effectively abolish binding, researchers have tested combinations of mutations at different positions. See, for example, Borrok 2017; Lin 2013; Vafa 2014; Schlothauer 2016.

[0097] The number of possible combinations comparable to the identified mutations (Table 3) is substantial (approximately 276,480). Clearly, it would be entirely impractical to generate and test all of the vast number of possible alternative substitution combinations at each site, let alone the even larger number of substitution combinations that may affect both C1q binding and FcγR binding. In practice, only a small number of combinations have been tested. Of these, very few have been used in therapeutic antibodies entering clinical trials, as exemplified in Table 2.

[0098] Measurement of binding to C1q A variety of different techniques have been used to measure the binding of variant immunoglobulins to C1q. Examples include enzyme-linked immunosorbent assays (ELISAs) (e.g., Hezarah 2001; Dall'Acqua 2006; Oganesyan 2008; Schlothauer 2016), flow cytometry (e.g., Kanda 2006), multiplex microsphere (Luminex) assays (e.g., Boesch 2014), and surface plasmon resonance (e.g., Borrock 2017; Dall'Acqua 2006; Moore 2010). Other experimental techniques and implementations of the aforementioned techniques are also known in the art. The wide variety of experimental techniques, experimental conditions, and various preparations of variant immunoglobulins and Fc receptors makes it impossible to reliably compare results from different studies to determine whether one particular combination of mutations is actually more effective than another for silencing binding to C1q.

[0099] Variant Fc regions with reduced binding to human FcγR and human C1q Comparison of Fc variants It is not possible to reliably compare results from different published studies to determine whether one particular combination of mutations is actually more effective than another for silencing binding to FcγR or C1q. Therefore, as described in Example 1, we prepared a set of variant immunoglobulins representing variants known in the art, but all based on the same IgG1 antibody (Table 4). Binding to human FcγR was measured by surface plasmon resonance as described in Example 2. Surprisingly, all of the previously known variants, even those previously described as having FcγR binding "completely eliminated or reduced to undetectable levels" (WO2013165690), "completely inactive" (WO2014108483), "completely abolished" (Schlothauer 2016; Hezarah 2001), "silent" (Tam 2017), or "no detectable binding" (Vafa 2014), conferred measurable binding to human FcγRI with responses significantly above background (Table 8). Furthermore, as shown in Example 6, some of the previously known variants, including L234A / L235A, L234F / L235E / P331S, and L234F / L235Q / K322Q, also conferred responses significantly above background in ADCC and ADCP assays (Tables 20 and 21).

[0100] [Table 5]

[0101] Characterization of new Fc variants We now prepared a novel set of Fc variants as described in Example 1 and compared them with known variants. From the results, we were able to identify a small number of amino acid residues whose amino acid substitutions could be combined to result in variant Fc regions with significantly lower levels of binding to human FcγRI than known variants. We identified variant Fc regions containing two or three amino acid substitutions that substantially reduced or completely abolished detectable binding to human Fcγ receptors as measured using surface plasmon resonance (see, e.g., Example 2) or, for ADCC or ADCP, by cell-based assays (see, e.g., Example 6). In most cases, the same amino acid substitutions also substantially reduced or completely abolished detectable binding to human C1q as measured using ELISA (see, e.g., Example 4). In some of the experimental tests to be described, proteins containing variant Fc regions containing amino acid substitutions will be compared to blank samples containing buffer only. In other tests, proteins containing variant Fc regions containing amino acid substitutions will be compared to a reference protein. In such cases, the proteins to be compared must be produced using essentially the same cell line and culture conditions, and purified by essentially the same method, so that the results of the comparative test can be valid. Preferably, all components of the test protein and reference protein, except for the Fc region, are essentially identical. For example, if the test protein is an antibody, both the test protein and the reference protein preferably have the same variable region. Depending on the purpose of the test, the reference protein may contain a naturally occurring (wild-type) Fc region (e.g., SEQ ID NO: 1) or a variant Fc region having amino acid changes corresponding to the previously described Fc variants, preferably L234A / L235A / P329G (LALAPG) (e.g., SEQ ID NO: 2).

[0102] The binding of the new variants to human FcRn is not substantially reduced compared to the binding of the wild-type reference protein, as measured by surface plasmon resonance (see, e.g., Example 3). Furthermore, the thermal stability of the new variants is not substantially reduced compared to the thermal stability of the wild-type reference protein, as measured by size exclusion chromatography (see, e.g., Example 7), or by differential scanning fluorimetry (see, e.g., Examples 8 and 12), or by light scattering (see, e.g., Example 12) after incubation at 40°C for up to 14 days.

[0103] Proteins containing variant Fc regions with significantly reduced binding activity Thus, in some embodiments, a) an amino acid substitution at position 234 or an amino acid substitution at position 235 or an amino acid substitution at both positions 234 and 235, and b) Amino acid change to arginine (R) at position 236 1. A protein comprising a variant Fc region comprising: Proteins are provided in which the binding of the protein to human FcγRI is significantly reduced compared to the binding of a reference protein, the reference protein comprising a variant human IgG1 Fc region comprising amino acid substitutions L234A, L235A and P329G.

[0104] An "amino acid change" at position 236 means either an amino acid substitution (eg, in the case of an Fc region from human IgG1, IgG3, IgG4) or an amino acid insertion (eg, in the case of an Fc region from human IgG2).

[0105] In some embodiments, a) an amino acid substitution at position 234 or at both positions 234 and 235; and b) Amino acid change to arginine (R) at position 236 1. A protein comprising a variant Fc region comprising: The binding of the protein to human FcγRI is significantly reduced compared to the binding of a reference protein, the reference protein comprising a variant human IgG1 Fc region comprising amino acid substitutions L234A, L235A and P329G. is provided.

[0106] In all of the following embodiments of the above aspects, the protein comprising the variant Fc region comprises an amino acid change at position 236 to arginine (R), and binding of the protein to human FcγRI is significantly reduced compared to binding of the reference protein.

[0107] In some embodiments, the amino acid at position 234 is substituted with any one of A, D, E, G, H, K, Q, R, S, or T. In some embodiments, the amino acid at position 235 is substituted with any one of A, D, E, G, H, I, K, Q, S, T, or V.

[0108] In some embodiments, the amino acid at position 234 is substituted with any one of A, D, E, G, H, K, Q, R, S, or T, and the amino acid at position 235 is substituted with any one of A, D, E, G, H, I, K, Q, R, S, T, or V.

[0109] Some amino acid substitutions or alterations result in a change in charge. For example, substitution of an uncharged residue (e.g., A, F, G, L) with a negatively charged residue (e.g., D, E) results in a net increase in negative charge. Conversely, substitution with a positively charged residue (e.g., H, K, R) results in a net increase in positive charge. Substitution with a negatively charged residue at one site and a positively charged residue at another site is not expected to result in a change in the net number of charged residues. Other combinations will be apparent to those skilled in the art. A change in the net number of charged residues affects the isoelectric point of the protein, which may have adverse effects on its ease of manufacture, stability, solubility, in vivo half-life, or other physicochemical or biological properties. It is desirable to minimize such changes. Therefore, in some embodiments of this invention, the amino acid substitutions and alterations at any of positions 234, 235, and 236 result in no more than one net change in charged amino acid residues. For example, in some embodiments, the amino acid changes at positions 234, 235, and 236 may be selected from 0 / 0 / +, 0 / - / +, - / 0 / +, - / - / +, - / +, or + / - / +, where "0" indicates either no change or a substitution with an uncharged amino acid, "-" indicates a substitution with a negatively charged amino acid, and "+" indicates a substitution with a positively charged amino acid. Less desirable embodiments, which are expected to change the net number of charged residues by 2 or 3, include 0 / + / +, + / 0 / +, and + / + / +. In some embodiments, any substitution with a positively charged amino acid at positions 234 or 235 is avoided.

[0110] In some embodiments, a protein is provided comprising a variant Fc region, wherein the variant Fc region comprises (a) an amino acid substitution at position 234 or an amino acid substitution at position 235, or an amino acid substitution at both positions 234 and 235; and (b) an amino acid alteration at position 236 to arginine (R), where amino acid numbering is according to the EU index as set forth in Kabat, wherein binding of the protein to human FcγRI is significantly reduced compared to binding of a LALAPG reference protein, and wherein said amino acid substitutions and alterations result in no more than one net change in charged amino acid residues.

[0111] In some embodiments, the amino acid at position 234 is substituted with any one of H, K, or R and the amino acid at position 235 is substituted with any one of D or E, or the amino acid at position 234 is substituted with any one of D or E and the amino acid at position 235 is substituted with any one of H, K, or R. In some embodiments, the amino acid at position 234 is substituted with any one of A, D, E, G, Q, S, or T. In some embodiments, the amino acid at position 235 is substituted with any one of A, D, E, G, I, Q, S, T, or V.

[0112] In some embodiments, the amino acid at position 234 is unchanged and the amino acid at position 235 is substituted with any one of E, Q, S, or T.

[0113] In some embodiments, the binding of a protein comprising a variant Fc region to human FcγRI is less than 50%, preferably less than 40%, less than 30%, less than 20%, less than 10%, or less than 5% of the binding of a reference protein comprising a variant human IgG1 Fc region comprising amino acid substitutions L234A, L235A, and P329G. Binding of a protein to human FcγRI may be measured by any suitable method sensitive enough to detect small differences in binding. Preferably, binding is measured by surface plasmon resonance, e.g., as described in Example 2. In some embodiments, the protein comprising the variant Fc region has a binding response to human FcγRI of less than 40 response units (RU), preferably less than 30 RU, less than 20 RU, less than 10 RU, or less than 5 RU, as measured by surface plasmon resonance, under conditions where an equivalent wild-type protein produces a binding response of greater than 2000 RU, or where a reference protein comprising a variant human IgG1 Fc region comprising amino acid substitutions L234A, L235A, and P329G produces a binding response of greater than 40 RU, as measured by surface plasmon resonance.

[0114] In some embodiments, a protein comprising a variant Fc of a wild-type human IgG1 Fc region, a) an amino acid substitution at position 234 or an amino acid substitution at position 235 or an amino acid substitution at both positions 234 and 235, and b) Amino acid substitution at position 236 to arginine (R) Including, Proteins are provided in which the binding of the protein to human FcγRI is significantly reduced compared to the binding of a reference protein, the reference protein comprising a variant human IgG1 Fc region comprising amino acid substitutions L234A, L235A and P329G.

[0115] In some embodiments, a protein comprising a variant Fc of a wild-type human IgG1 Fc region, a) an amino acid substitution at position 234 or at both positions 234 and 235, and b) Amino acid substitution at position 236 to arginine (R) Including, Proteins are provided in which the binding of the protein to human FcγRI is significantly reduced compared to the binding of a reference protein, the reference protein comprising a variant human IgG1 Fc region comprising amino acid substitutions L234A, L235A and P329G.

[0116] In some embodiments, provided are proteins comprising a variant Fc region, wherein the variant Fc region comprises (a) an amino acid substitution at position 234 or an amino acid substitution at position 235, or an amino acid substitution at both positions 234 and 235; and (b) an amino acid alteration at position 236 to arginine (R), where amino acid numbering is according to the EU index as set forth in Kabat, wherein binding of the protein to human FcγRI is significantly reduced compared to binding of a LALAPG reference protein, and wherein said amino acid substitutions and alterations result in no more than one net change in charged amino acid residues.

[0117] In some embodiments, the binding of the protein to human C1q is effectively undetectable (i.e., not significantly greater than the binding of a sample containing buffer alone). In some embodiments, the binding of the protein to human C1q is substantially reduced compared to a comparable wild-type reference protein. In some embodiments, the binding of the protein to human C1q is less than 20%, preferably less than 10%, less than 5%, less than 2%, or less than 1% of the binding of the wild-type reference protein. In some embodiments, the binding of the protein to human C1q is not significantly different from the binding of the LALAPG reference protein. The binding of the protein to human C1q may be measured by any suitable method that is sensitive enough to detect small differences in binding. Preferably, the binding is measured by ELISA, for example, as described in Example 4. In some embodiments, a protein comprising a variant Fc region has a specific binding response to human C1q, as measured by ELISA, of less than 0.1 absorbance units (AU), preferably less than 0.05 AU, less than 0.02 AU, less than 0.01 AU, or less than 0.015 AU, under conditions under which the equivalent wild-type protein produces a specific binding response of greater than 1.0 AU, as measured by ELISA, where a "specific binding response" is the binding response of the test sample that is lower than the binding response of buffer alone.

[0118] In some embodiments, proteins comprising a variant Fc region retain human FcRn binding, and in some embodiments, FcRn binding is reduced by 50% or less, 40% or less, 30% or less, 20% or less, 10% or less, or 5% or less compared to the wild-type reference protein. In some embodiments, there is no significant difference between the binding of a protein comprising a variant Fc region to human FcRn and the binding of the wild-type reference protein. In some embodiments, a protein comprising a variant Fc region has greater binding activity to human FcRn than the wild-type reference protein. Binding of a protein to human FcRn may be measured by any suitable method sensitive enough to detect differences in binding, such as by surface plasmon resonance as described in Example 3 or by a NanoBiT® competitive immunoassay as described in Example 3.

[0119] Proteins containing variant Fc regions with substantially reduced ADCC In some embodiments, proteins comprising a variant Fc region exhibit substantially reduced ADCC activity compared to the corresponding wild-type reference protein. In some embodiments, the ADCC activity of a protein comprising a variant Fc region is less than 20%, preferably less than 10%, less than 5%, less than 2%, or less than 1% of the ADCC activity of the corresponding wild-type reference protein. In some embodiments, the ADCC activity of a protein comprising a variant Fc region is not significantly different from the ADCC activity of assay buffer. In some embodiments, a protein comprising a variant Fc region has no detectable ADCC activity at concentrations up to 10 μg / mL. In some embodiments, the ADCC activity of a protein comprising a variant Fc region is not significantly different from the ADCC activity of a LALAPG reference protein comprising amino acid substitutions L234A, L235A, and P329G. In some embodiments, the ADCC activity of a protein comprising a variant Fc region is significantly lower than the ADCC activity of a reference protein comprising amino acid substitutions L234A and L235A. In some embodiments, the ADCC activity of a protein comprising a variant Fc region is less than 50%, preferably less than 20%, less than 10%, less than 5%, less than 2%, or less than 1% of the ADCC activity of a reference protein comprising amino acid substitutions L234A and L235A. ADCC activity may be measured by any suitable method sensitive enough to detect differences in activity, for example, by a cell-based luminescence assay such as that described in Example 6.

[0120] Proteins containing variant Fc regions with substantially reduced ADCP In some embodiments, a protein comprising a variant Fc region exhibits substantially reduced ADCP activity compared to a corresponding wild-type reference protein. In some embodiments, the ADCP activity of a protein comprising a variant Fc region is less than 20%, preferably less than 10%, less than 5%, less than 2%, or less than 1% of the ADCP activity of the corresponding wild-type reference protein. In some embodiments, the ADCP activity of a protein comprising a variant Fc region is not significantly different from the ADCP activity of an assay buffer. In some embodiments, a protein comprising a variant Fc region has no detectable ADCP activity at concentrations up to 10 μg / mL. In some embodiments, the ADCP activity of a protein comprising a variant Fc region is not significantly different from the ADCP activity of a reference protein comprising amino acid substitutions L234A, L235A, and P329G. In some embodiments, the ADCP activity of a protein comprising a variant Fc region is significantly lower than the ADCP activity of a reference protein comprising amino acid substitutions L234A and L235A. In some embodiments, the ADCP activity of a protein comprising a variant Fc region is less than 50%, preferably less than 20%, less than 10%, less than 5%, less than 2%, or less than 1% of the ADCP activity of a reference protein comprising amino acid substitutions L234A and L235A. ADCP activity may be measured by any suitable method sensitive enough to detect differences in activity, for example, by a cell-based luminescence assay such as that described in Example 6.

[0121] Proteins containing variant Fc regions with substantially reduced CDC In some embodiments, proteins comprising a variant Fc region exhibit substantially reduced CDC activity compared to the corresponding wild-type reference protein. In some embodiments, the CDC activity of a protein comprising a variant Fc region is less than 20%, preferably less than 10%, less than 5%, less than 2%, or less than 1% of the CDC activity of the corresponding wild-type reference protein. In some embodiments, the CDC activity of a protein comprising a variant Fc region is not significantly different from the CDC activity of the assay buffer. In some embodiments, a protein comprising a variant Fc region has no detectable CDC activity at concentrations up to 10 μg / mL. In some embodiments, the CDC activity of a protein comprising a variant Fc region is not significantly different from the CDC activity of a reference protein comprising the amino acid substitutions L234A, L235A, and P329G. CDC activity may be measured by any suitable method sensitive enough to detect differences in activity, for example, 51 It may be measured by Cr release assay (e.g., Hale 1983) or luminescence assay (e.g., Niles 2007).

[0122] Proteins containing variant Fc regions with substantially reduced toxicity Certain proteins containing Fc regions, particularly certain monoclonal antibodies (e.g., muromonab, alemtuzumab, and ceralizumab), are known to induce toxicological effects in humans associated with the release of proinflammatory cytokines (e.g., gamma interferon, tumor necrosis factor, and interleukin-6). Such toxicological effects (sometimes referred to as "cytokine storm") are often caused by or can be exacerbated by the binding of Fc regions to Fcγ receptors. Various in vitro assays that predict such toxicological effects (to varying degrees) are known in the art (e.g., Finco 2014; Grimaldi 2016; Vessilier 2015). For example, in so-called "whole blood" assays, a test sample is incubated with unfractionated blood (preferably anticoagulated with heparin) and the release of proinflammatory cytokines is measured (e.g., Wing 1995; Wolf 2012). Alternatively, the test sample may be incubated with peripheral blood mononuclear cells (PBMCs) and the release of inflammatory cytokines measured (e.g., Vessilier 2015). Inflammatory cytokines may include some or all of GM-CSF, IFNγ, IL-2, IL-4, IL-6, IL-8, IL-10, TNFα, or other cytokines known to be associated with inflammatory responses. In some assays, the test sample may be immobilized, for example, by binding to a microplate (e.g., Findlay 2010) or onto epithelial cells (e.g., Findlay 2011). In other assays, sensitivity can be increased by pre-culturing PBMCs at high density (e.g., Romer 2011). Cytokine release activity may be measured by determining the concentration of one or more cytokines in the culture supernatant of cells exposed to the test sample for a period of 1 to 72 hours. Cytokine release activity may be measured using a single donor, more than one donor, or averaged across a number of donors. In some embodiments, proteins comprising a variant Fc region exhibit substantially reduced cytokine release activity compared to a wild-type reference protein.In some embodiments, the cytokine release activity of the protein comprising a variant Fc region is less than 20%, preferably less than 10%, less than 5%, less than 2%, or less than 1% of the cytokine release activity of the corresponding wild-type reference protein. In some embodiments, the cytokine release activity of the protein comprising a variant Fc region is lower than or not significantly different from the cytokine release activity of the diluent. In some embodiments, the cytokine release activity of the protein comprising a variant Fc region is not significantly different from the cytokine release activity of the LALAPG reference protein. Cytokine release activity may be measured by any suitable method sensitive enough to detect differences in activity, for example, as described in Example 9.

[0123] Proteins containing variant Fc regions with minimal immunogenicity Key steps in the immune response to proteins include proteolytic processing and binding of peptides to major histocompatibility complex (MHC) class II for presentation on the surface of antigen-presenting cells. Such peptide-MHC complexes can be recognized by specific receptors on CD4+ T cells, resulting in T cell activation and B cell support for antibody responses. The propensity to generate new peptides with a high probability of binding to MHC class II can be determined by a variety of methods, including both "in silico" methods (e.g., Jensen 2018; Nielson 2007; Sidney 2008; Sturniolo 1999; Bryson 2010; Jawa 2013; King 2014) or "in vitro" methods (e.g., Brinks 2013; Jawa, 2013; Joubert 2016).

[0124] In some embodiments, a protein comprising a variant Fc region comprises an amino acid substitution that does not have the property of generating new peptides with a high probability of binding to human MHC class II, as determined by "in silico" methods. In some embodiments, a protein comprising a variant Fc region does not contain any new peptides (compared to the wild-type reference protein) with a rank score of ≦10% when assessed using the IEDB MHC-II binding prediction tool (Fleri 2017). Examples are provided in Example 5. Amino acid substitutions that generate new peptides with a higher probability of binding to human MHC class II include L234F, L234I, L234M, L234W, L234Y, L235F, L235W, and L235Y (see Table 19). In some embodiments, the protein comprises an amino acid substitution at residue 234 that does not include phenylalanine (F), isoleucine (I), methionine (M), tryptophan (W), or tyrosine (Y). In some embodiments, the protein comprises an amino acid substitution at residue 235 that does not include phenylalanine (F), tryptophan (W), or tyrosine (Y). In some embodiments, the protein comprises an amino acid substitution that does not include any of the above substitutions.

[0125] In some embodiments, the protein contains an amino acid substitution that does not have the property of creating new peptides with a high probability of being presented on the surface of human antigen-presenting cells (e.g., dendritic cells). The property of creating new peptides with a high probability of being presented on the surface of human antigen-presenting cells can be determined by any suitable method (e.g., Brinks 2013; Jawa 2013; Joubert 2016). An example is provided in Example 13. Surprisingly, a peptide with only the amino acid substitution G236R was found to have an increased risk of immunogenicity compared to the corresponding wild-type peptide, whereas adding additional substitutions at positions 234 and 235 reduced this risk to below that of the corresponding wild-type peptide.

[0126] In some embodiments, proteins comprising variant Fc regions have significantly reduced immunogenicity in vivo compared to the corresponding wild-type reference protein. In vivo immunogenicity can be measured by any suitable method known in the art. For example, a subject, such as a mouse, rat, rabbit, non-human primate, or human, may be immunized with one or more doses of a test sample. After a suitable period of time (e.g., 7-28 days) after immunization, cellular and / or humoral immune responses may be measured (e.g., Chen 2007; Loureiro 2011). Preferably, a sufficient number of subjects (e.g., five or more) are immunized with each test substance to allow statistical comparison of measured immune responses between groups immunized with different proteins.

[0127] Proteins containing variant Fc regions with unchanged glycosylation Immunoglobulin Fc regions expressed by mammalian cells are typically glycosylated at N297 and exhibit a spectrum of different glycoforms characteristic of the production process, as well as the selection of host cells and cell culture conditions, among others (e.g., Jefferis 2005; Costa 2014; Werner 2007). Numerous methods are known in the art for characterizing the spectrum of glycoforms exhibited by proteins comprising Fc regions (see, e.g., Reutsch 2015a; Reutsch 2015b). In some embodiments, the spectrum of glycoforms exhibited by proteins comprising variant Fc regions is indistinguishable from the spectrum of glycoforms exhibited by the corresponding wild-type reference protein. In some embodiments, the spectrum of glycoforms exhibited by proteins comprising variant Fc regions is not significantly different from the spectrum of glycoforms exhibited by the corresponding wild-type reference protein. In some embodiments, the spectrum of glycoforms exhibited by proteins comprising variant Fc regions is not substantially different from the spectrum of glycoforms exhibited by the corresponding wild-type reference protein. In each case, the proteins being compared are produced using essentially the same cell line and essentially the same culture conditions, and are purified in essentially the same manner. An example is provided in Example 11.

[0128] Proteins containing variant Fc regions are homogeneous and stable during production and storage Proteins intended for use in pharmaceutical production are desirably homogeneous and stable over long periods of time. Certain amino acid residues, including asparagine (N), cysteine ​​(C), and methionine (M), can be subject to post-translational modifications, such as deamidation of asparagine or oxidation of cysteine ​​or methionine, which can lead to undesirable heterogeneity or instability. Motifs particularly prone to deamidation include NG, NS, NH, and QG. Other amino acid substitutions, such as glycine (G) and proline (P), can destabilize the protein backbone and disrupt the three-dimensional structure. Motifs prone to aspartic acid isomerization include DG, DP, and DS. Motifs prone to peptide bond cleavage include TS, KK, RK, and KR. In some embodiments, proteins contain amino acid substitutions that do not include asparagine (N), cysteine ​​(C), or methionine (M). In some embodiments, proteins contain amino acid substitutions that do not include glycine (G) or proline (P). In some embodiments, the protein comprises amino acid substitutions that do not create a two amino acid motif, such as NG, NS, NH, QG, DG, DP, DS, TS, KK, RK, or KR.

[0129] Various amino acid substitutions can increase heterogeneity or instability by increasing the tendency of a protein to unfold or form less ordered forms. Such tendency can be determined by various techniques known in the art, including measuring the thermal stability of a protein using, for example, differential scanning calorimetry (DSC), differential scanning fluorometry (DSF), circular dichroism (CD), differential static light scattering (DSLS), isothermal denaturation (ITD), or a thermal challenge assay (see, e.g., Senisterra 2009). Examples are provided in Examples 8 and 12. Other techniques that may be particularly suitable for the rapid measurement of small sample volumes include affinity-capture self-interaction nanoparticle spectroscopy (AC-SINS), clonal self-interaction by biolayer interferometry (CSI-BLI), hydrophobic interaction chromatography (HIC), salt gradient affinity capture self-interaction nanoparticle spectroscopy (SGAC-SINS), and standup monolayer adsorption chromatography (e.g., Jain 2017 and references therein).

[0130] In some embodiments, the thermal stability of the protein comprising a variant Fc region is not significantly different from the thermal stability of an equivalent wild-type reference protein. In some embodiments, the thermal stability of the protein comprising a variant Fc region is within ±3°C, ±2°C, or ±1°C of the thermal stability of the equivalent wild-type reference protein. In some embodiments, the thermal stability of the protein comprising a variant Fc region is greater than the thermal stability of the equivalent wild-type reference protein.

[0131] Various amino acid substitutions can increase a protein's tendency to form dimers or higher-order aggregates. Such tendency can be determined by various techniques known in the art, such as size exclusion chromatography (SEC), ultracentrifugation, dynamic light scattering (DLS), and many others (see, e.g., Engelsman 2011). In some embodiments, the proportion of monomers in a protein comprising a variant Fc region is within ±3%, ±2%, or ±1% of the proportion of monomers in an equivalent wild-type reference protein. In some embodiments, the proportion of monomers in a protein comprising a variant Fc region is not significantly different from the proportion of monomers in an equivalent wild-type reference protein. In some embodiments, the proportion of monomers in a protein comprising a variant Fc region is greater than the proportion of monomers in an equivalent wild-type reference protein.

[0132] A biopharmaceutical product is considered stable as long as its characteristics remain within the manufacturer's specifications, despite changes over time. The number of days a product remains stable under recommended storage conditions is referred to as its shelf life. Experimental protocols commonly used for collecting data to serve as a basis for estimating a product's shelf life are called stability studies. Stability testing is generally performed by means of real-time stability studies under recommended storage conditions or accelerated stability studies under elevated stress conditions, such as protein concentration, temperature, humidity, shaking, or light exposure (see, e.g., Bajaj 2012). In some embodiments, accelerated stability studies can be performed at one or more protein concentrations ranging from 1 mg / mL to 100 mg / mL. In some embodiments, accelerated stability studies can be performed at one or more temperatures ranging from 25°C to 60°C. In some embodiments, the stability of a protein comprising a variant Fc region is within ±20%, ±10%, ±5%, or ±2% of the stability of an equivalent wild-type reference protein. In some embodiments, the stability of a protein comprising a variant Fc region is not significantly different from that of an equivalent wild-type reference protein. In some embodiments, the stability of a protein comprising a variant Fc region is greater than that of an equivalent wild-type reference protein. In each case, stability may be measured using any one or more of any suitable stability studies or tests, including real-time stability studies, accelerated stability studies, at any suitable protein concentration or temperature, using any suitable experimental measurement, such as DSC, DSF, CD, thermal challenge assay, SEC, ultracentrifugation, DLS, or measuring binding activity to an antigen or other ligand or receptor, among others (see, e.g., Senisterra 2009; Jain 2017). Those skilled in the art will appreciate the many different protocols for implementing these experimental techniques. For example, thermal stability may be measured by DSF using the fluorescent probe SYPRO® Orange, although a variety of other dyes can also be used (e.g., Niesen 2007).Or, as a further example, binding to an antigen or other ligand or receptor can be measured by surface plasmon resonance, enzyme-linked immunosorbent assay, NanoBiT® immunoassay, or many other types of ligand binding assays. Examples of stability measurements are provided in Examples 7, 8, and 12.

[0133] Proteins containing variant Fc regions with unchanged pharmacokinetics Proteins containing an Fc region generally have an extended half-life in vivo due to binding to the FcRn receptor. Various methods for measuring in vivo half-life are well known in the art (see, e.g., Liu 2018). Typically, these involve injecting the protein into a suitable animal (e.g., a mouse or rat), collecting blood samples at suitable intervals, measuring the protein concentration in the plasma (or serum), and mathematically analyzing the resulting data. In addition to half-life, other pharmacokinetic parameters may be determined, such as volume of distribution, maximum plasma concentration, area under the curve, and clearance.

[0134] In some embodiments, the pharmacokinetic parameters exhibited by a protein comprising a variant Fc region are indistinguishable from the pharmacokinetic parameters exhibited by the corresponding wild-type reference protein. In some embodiments, the pharmacokinetic parameters exhibited by a protein comprising a variant Fc region are not significantly different from the pharmacokinetic parameters exhibited by the corresponding wild-type reference protein. In some embodiments, the pharmacokinetic parameters exhibited by a protein comprising a variant Fc region are not substantially different from the pharmacokinetic parameters exhibited by the corresponding wild-type reference protein. In some embodiments, the protein comprising a variant Fc region differs in one or more pharmacokinetic parameters from the half-life of the wild-type reference protein by less than 50%, preferably less than 20%, less than 10%, or less than 5%. In each case, the proteins being compared are produced using essentially the same cell line and essentially the same culture conditions, and purified by essentially the same method. In each case, "pharmacokinetic parameters" means any one or more of the half-life, volume of distribution, maximum plasma concentration, area under the curve, and clearance rate, determined using appropriate mathematical methods from concentration measurements in blood, plasma, or serum samples after administration of the protein to a mammal.

[0135] In some embodiments, pharmacokinetic parameters are measured after administering the test sample to mice transgenic for human FcRn, eg, as described in Example 10.

[0136] Methods for reducing Fc-induced effector functions In some embodiments, methods are provided for substantially reducing Fc-induced effector function of a protein comprising an Fc region by substantially reducing any one or more of the following: binding to one or more Fcγ receptors, binding to C1q, ADCC, ADCP, CDC. The methods include: a) substituting the amino acid at position 234 or the amino acid at position 235 or the amino acids at both positions 234 and 235; and b) changing the amino acid at position 236 to arginine (R) wherein binding of the protein to human FcγRI is significantly reduced compared to binding of a reference protein comprising the amino acid substitutions L234A, L235A, and P329G. In some embodiments, proteins comprising the resulting variant Fc region have one or more of the desirable properties listed in more detail above: (a) similar or greater binding activity to FcRn compared to the wild-type reference protein, (b) substantially reduced ADCC compared to the wild-type reference protein, (c) substantially reduced CDC compared to the wild-type reference protein, (d) substantially reduced toxicity compared to the wild-type reference protein, (e) minimal immunogenicity compared to the wild-type reference protein, (f) unchanged glycosylation compared to the wild-type reference protein, (g) similar or improved homogeneity and / or stability during manufacturing and / or storage compared to the wild-type reference protein, and (h) similar pharmacokinetic parameters compared to the wild-type reference protein.

[0137] The necessary amino acid substitutions are achieved using any of a variety of techniques well known in the art. Typically, these involve (a) synthesizing and / or assembling DNA encoding a protein with the desired amino acid substitutions, (b) expressing the protein encoded by the DNA using a suitable cell or non-cell-based system, (c) isolating the protein, and (d) testing the isolated protein for the desired reduction in Fc-induced effector function. Examples of steps (a), (b), and (c) are provided in Example 1. Examples of step (d) are provided in Examples 2, 4, 5, 6, and 9. Many possible variations on this method will be apparent to those skilled in the art.

[0138] Proteins containing variant Fc regions with specific amino acid substitutions As shown in Example 2 and Table 8, the inventors surprisingly identified proteins containing variant Fc regions with novel sets of amino acid substitutions that result in significantly lower binding to FcγRI than that seen in the equivalent LALAPG reference protein (despite previously being reported to have "completely abolished" FcγR and C1q interactions; see Schlothauer 2016). Any of these novel sets of amino acid substitutions can be used to reduce binding to FcγRI compared to the LALAPG reference.

[0139] In some embodiments, L234A / L235A / G236R, L234A / L235S / G236R, L234A / L235T / G236R, L234D / L235H / G236R, L234D / L235K / G236R, L234D / L235Q / G236R, L234D / L235S / G236R, L234D / L235T / G236R, L234E / L235D / G236R, L234E / L235H / G236R, L234E / L235I / G236R, L234E / L235V / G236R, L234G / L235H / G236R, L234G / L235Q / G236R, L234G / L235S / G236R, L234H / L235I / G236R, L234H / L235S / G236R, L234K / L235Q / G236R, L234K / L235R / G236R, L234K / L235S / G236R, L234K / L235T / G236R, L234K / L235V / G236R, L234Q / L235A / G236R, L234Q / L235D / G236R, L234Q / L235H / G236R, L234Q / L235Q / G236R, L234Q / L235R / G236R, L234Q / L235S / G236R, L234Q / L235T / G236R, L234Q / L235V / G236R, L234R / L235D / G236R, L234R / L235E / G236R, L234R / L235H / G236R, L234R / L235I / G236R, L234R / L235K / G236R, L234R / G236R, L234R / L235Q / G236R, L234R / L235R / G236R, L234R / L235T / G236R, L234S / L235D / G236R, L234S / L235E / G236R, L234S / L235G / G236R, L234S / L235H / G236R, L234S / L235I / G236R, L234S / G236R, L234S / L235R / G236R, L234S / L235T / G236R, L234S / L235V / G236R, L234T / L235A / G236R, L234T / L235I / G236R, L234T / L235K / G236R, L234T / L235Q / G236R, L234T / L235R / G236R, L234T / L235S / G236R, L234T / L235T / G236RProteins are provided that include a variant Fc region that includes a set of amino acid substitutions selected from L234T / L235V / G236R.

[0140] In some embodiments, L234A / L235A / G236R, L234A / L235S / G236R, L234A / L235T / G236R, L234D / L235H / G236R, L234D / L235K / G236R, L234D / L235Q / G236R, L234D / L235S / G236R, L234D / L235T / G236R, L234E / L235D / G236R, L234E / L235H / G236R, L234E / L235I / G236R, L234E / L235V / G236R, L234G / L235H / G236R, L234G / L235Q / G236R, L234G / L235S / G236R, L234H / L235I / G236R, L234H / L235S / G236R, L234K / L235Q / G236R, L234K / L235R / G236R, L234K / L235S / G236R, L234K / L235T / G236R, L234K / L235V / G236R, L234Q / L235A / G236R, L234Q / L235D / G236R, L234Q / L235H / G236R, L234Q / L235Q / G236R, L234Q / L235R / G236R, L234Q / L235S / G236R, L234Q / L235T / G236R, L234Q / L235V / G236R, L234R / L235D / G236R, L234R / L235E / G236R, L234R / L235H / G236R, L234R / L235I / G236R, L234R / L235K / G236R, L234R / G236R, L234R / L235Q / G236R, L234R / L235R / G236R, L234R / L235T / G236R, L234S / L235D / G236R, L234S / L235E / G236R, L234S / L235G / G236R, L234S / L235H / G236R, L234S / L235I / G236R, L234S / G236R, L234S / L235R / G236R, L234S / L235T / G236R, L234S / L235V / G236R, L234T / L235A / G236R, L234T / L235I / G236R, L234T / L235K / G236R, L234T / L235Q / G236R, L234T / L235R / G236R, L234T / L235S / G236R, L234T / L235T / G236R,Proteins are provided that include a variant Fc region that includes a set of amino acid substitutions selected from L234T / L235V / G236R, L235T / G236R.

[0141] In some embodiments, L234A / L235A / G236R, L234A / L235S / G236R, L234A / L235T / G236R, L234D / L235H / G236R, L234D / L235K / G236R, L234D / L235Q / G236R, L234D / L235S / G236R, L234D / L235T / G236R, L234E / L235D / G236R, L234E / L235H / G236R, L234E / L235I / G236R, L234E / L235V / G236R, L234G / L235Q / G236R, L234G / L235S / G2 36R, L234Q / L235A / G236R, L234Q / L235D / G236R, L234Q / L235Q / G236R, L234Q / L235S / G236R, L234Q / L235 T / G236R, L234Q / L235V / G236R, L234R / L235D / G236R, L234R / L235E / G236R, L234S / L235D / G236R, L234S / L235E / G236R, L234S / L235G / G236R, L234S / L235I / G236R, L234S / G236R, L234S / L235T / G236R, L234S / L2 Proteins are provided that comprise a variant Fc region comprising a set of amino acid substitutions selected from L234T / L235A / G236R, L234T / L235I / G236R, L234T / L235Q / G236R, L234T / L235S / G236R, L234T / L235T / G236R, L234T / L235V / G236R, L235T / G236R.

[0142] In some embodiments, L234A / L235A / G236R, L234A / L235S / G236R, L234D / L235K / G236R, L234D / L235S / G236R, L234D / L235T / G236R, L234G / L235S / G236R, L234H / L235S / G236R, L234K / L235Q / G236R, L234K / L235R / G236R, L234K / L 235S / G236R, L234K / L235T / G236R, L234K / L235V / G236R, L234Q / L235A / G236R, L234Q / L235D / G236R, L234Q / L 235R / G236R, L234Q / L235S / G236R, L234Q / L235T / G236R, L234Q / L235V / G236R, L234R / L235D / G236R, L234R / L 235E / G236R, L234R / L235H / G236R, L234R / L235I / G236R, L234R / L235K / G236R, L234R / L235Q / G236R, L234R / L 235R / G236R, L234S / L235G / G236R, L234S / L235H / G236R, L234S / L235I / G236R, L234S / L235R / G236R, L234S / L Proteins are provided that comprise a variant Fc region comprising a set of amino acid substitutions selected from 235T / G236R, L234S / L235V / G236R, L234T / L235K / G236R, L234T / L235Q / G236R, L234T / L235R / G236R, L234T / L235S / G236R, L234T / L235T / G236R, L234T / L235V / G236R.

[0143] In some embodiments, L234A / L235A / G236R, L234A / L235S / G236R, L234D / L235K / G236R, L234D / L235S / G236R, L234D / L235T / G236R, L234G / L235S / G236R, L234H / L235S / G236R, L234K / L235Q / G236R, L234K / L 235R / G236R, L234K / L235S / G236R, L234K / L235V / G236R, L234Q / L235A / G236R, L234Q / L235D / G236 R, L234Q / L235R / G236R, L234Q / L235S / G236R, L234Q / L235T / G236R, L234Q / L235V / G236R, L234R / L 235E / G236R, L234R / L235H / G236R, L234R / L235K / G236R, L234R / L235Q / G236R, L234R / L235R / G236 R, L234S / L235G / G236R, L234S / L235H / G236R, L234S / L235I / G236R, L234S / L235R / G236R, L234S / L Proteins are provided that comprise a variant Fc region comprising a set of amino acid substitutions selected from 235T / G236R, L234S / L235V / G236R, L234T / L235Q / G236R, L234T / L235R / G236R, L234T / L235S / G236R, L234T / L235T / G236R, L234T / L235V / G236R.

[0144] In some embodiments, L234A / L235S / G236R, L234G / L235S / G236R, L234Q / L235A / G236R, L234Q / L235S / G236R, L234Q / L235T / G236R, L234Q / L235V / G236R, L234S / L235G / G236R, L234S / L235I / G236 Proteins are provided that comprise a variant Fc region comprising a set of amino acid substitutions selected from: L234S / L235T / G236R, L234S / L235V / G236R, L234T / L235Q / G236R, L234T / L235S / G236R, L234T / L235T / G236R, L234T / L235V / G236R.

[0145] In some embodiments, proteins are provided that comprise a variant Fc region comprising a set of amino acid substitutions selected from L234G / L235S / G236R, L234S / L235T / G236R, L234S / L235V / G236R, L234T / L235Q / G236R, L234T / L235T / G236R.

[0146] IgG subclasses In some embodiments, the Fc region altered to create the variant Fc region can be selected from IgG1, IgG2, IgG3, or IgG4. In some embodiments, the Fc region is IgG1. In humans, the wild-type residues corresponding to L234, L235, and G236 (EU numbering) in IgG1 are V234, A235, and Δ236 in IgG2; L234, L235, and G236 in IgG3; and F234, L235, and G236 in IgG4. IgG2 has a deletion at position 236, which is associated with reduced binding to FcRn and reduced transplacental transport (Stapleton 2018). In some embodiments, an Arg residue inserted at position 236 restores FcRn binding of IgG2 to more similar binding to IgG1, thereby increasing the half-life of IgG2 in circulation. In other embodiments, retention of the deletion at 236 maintains the reduced binding of IgG2 to FcRn.

[0147] In some embodiments, L234A / L235A / G236R, L234A / L235S / G236R, L234A / L235T / G236R, L234D / L235H / G236R, L234D / L235K / G236R, L234D / L235Q / G236R, L234D / L235S / G236R, L234D / L235T / G236R, L234E / L235D / G236R, L234E / L235H / G236R, L234E / L235I / G236R, L234E / L235V / G236R, L234G / L235H / G236R, L234G / L235Q / G236R, L234G / L235S / G236R, L234H / L235I / G236R, L234H / L235S / G236R, L234K / L235Q / G236R, L234K / L235R / G236R, L234K / L235S / G236R, L234K / L235T / G236R, L234K / L235V / G236R, L234Q / L235A / G236R, L234Q / L235D / G236R, L234Q / L235H / G236R, L234Q / L235Q / G236R, L234Q / L235R / G236R, L234Q / L235S / G236R, L234Q / L235T / G236R, L234Q / L235V / G236R, L234R / LA protein is provided comprising a human IgG1 variant Fc region comprising a set of amino acid substitutions selected from L234T / L235V / G236R.

[0148] In some embodiments, L234A / L235A / G236R, L234A / L235S / G236R, L234A / L235T / G236R, L234D / L235H / G236R, L234D / L235K / G236R, L234D / L235Q / G236R, L234D / L235S / G236R, L234D / L235T / G236R, L234E / L235D / G236R, L234E / L235H / G236R, L234E / L235I / G236R, L234E / L235V / G236R, L234G / L235H / G236R, L234G / L235Q / G236R, L234G / L235S / G236R, L234H / L235I / G236R, L234H / L235S / G236R, L234K / L235Q / G236R, L234K / L235R / G236R, L234K / L235S / G236R, L234K / L235T / G236R, L234K / L235V / G236R, L234Q / L235A / G236R, L234Q / L235D / G236R, L234Q / L235H / G236R, L234Q / L235Q / G236R, L234Q / L235R / G236R, L234Q / L235S / G236R, L234Q / L235T / G236R, L234Q / L235V / G236R, L234R / L235D / G236R, L234R / L235E / G236R, L234R / L235H / G236R, L234R / L235I / G236R, L234R / L235K / G236R, L234R / G236R, L234R / L235Q / G236R, L234R / L235R / G236R, L234R / L235T / G236R, L234S / L235D / G236R, L234S / L235E / G236R, L234S / L235G / G236R, L234S / L235H / G236R, L234S / L235I / G236R, L234S / G236R, L234S / L235R / G236R, L234S / L235T / G236R, L234S / L235V / G236R, L234T / L235A / G236R, L234T / L235I / G236R, L234T / L235K / G236R, L234T / L235Q / G236R, L234T / L235R / G236R, L234T / L235S / G236R, L234T / L235T / G236RProteins are provided that include a human IgG1 variant Fc region that includes a set of amino acid substitutions selected from L234T / L235V / G236R, L235T / G236R.

[0149] In some embodiments, L234A / L235A / G236R, L234A / L235S / G236R, L234A / L235T / G236R, L234D / L235H / G236R, L234D / L235K / G236R, L234D / L235Q / G236R, L234D / L235S / G236R, L234D / L235T / G236R, L234E / L235D / G236R, L234E / L235H / G236R, L234E / L235I / G236R, L234E / L235V / G236R, L234G / L235Q / G236R, L234G / L235S / G23 6R, L234Q / L235A / G236R, L234Q / L235D / G236R, L234Q / L235Q / G236R, L234Q / L235S / G236R, L234Q / L235T / G236R, L234Q / L235V / G236R, L234R / L235D / G236R, L234R / L235E / G236R, L234S / L235D / G236R, L234S / L23 5E / G236R, L234S / L235G / G236R, L234S / L235I / G236R, L234S / G236R, L234S / L235T / G236R, L234S / L235V / Proteins are provided that comprise a human IgG1 variant Fc region comprising a set of amino acid substitutions selected from G236R, L234T / L235A / G236R, L234T / L235I / G236R, L234T / L235Q / G236R, L234T / L235S / G236R, L234T / L235T / G236R, L234T / L235V / G236R, L235T / G236R.

[0150] In some embodiments, L234A / L235A / G236R, L234A / L235S / G236R, L234D / L235K / G236R, L234D / L235S / G236R, L234D / L235T / G236R, L234G / L235S / G236R, L234H / L235S / G236R, L234K / L235Q / G236R, L234K / L235R / G236R, L234K / L 235S / G236R, L234K / L235T / G236R, L234K / L235V / G236R, L234Q / L235A / G236R, L234Q / L235D / G236R, L234Q / L2 35R / G236R, L234Q / L235S / G236R, L234Q / L235T / G236R, L234Q / L235V / G236R, L234R / L235D / G236R, L234R / L23 5E / G236R, L234R / L235H / G236R, L234R / L235I / G236R, L234R / L235K / G236R, L234R / L235Q / G236R, L234R / L235 R / G236R, L234S / L235G / G236R, L234S / L235H / G236R, L234S / L235I / G236R, L234S / L235R / G236R, L234S / L235T / G236R, L234S / L235V / G236R, L234T / L235K / G236R, L234T / L235Q / G236R, L234T / L235R / G236R, L234T / L235S / G236R, L234T / L235T / G236R, L234T / L235V / G236R.

[0151] In some embodiments, V234A / Δ236R, V234A / A235T / Δ236R, V234D / A235H / Δ236R, V234D / A235K / Δ236R, V234D / A235Q / Δ236R, V234D / A235S / Δ236R, V234D / A235T / Δ236R, V234E / A235D / Δ236R, V234E / A235H / Δ236R, V234E / A235I / Δ236R, V234E / A235V / Δ236R, V234G / A235H / Δ236R, V234G / A235Q / Δ236R, V23 4G / A235S / Δ236R, V234H / A235I / Δ236R, V234H / A235S / Δ236R, V234K / A23 5Q / Δ236R, V234K / A235R / Δ236R, V234K / A235S / Δ236R, V234K / A235T / Δ236 R, V234K / A235V / Δ236R, V234Q / A235A / Δ236R, V234Q / A235D / Δ236R, V234Q / A235H / Δ236R, V234Q / A235Q / Δ236R, V234Q / A235R / Δ236R, V234Q / A235S / Δ236R, V234Q / A235T / Δ236R, V234Q / A235V / Δ236R, V234R / A235D / Δ236R, V234R / A235E / Δ236R, V234R / A235H / Δ236R, V234R / A235I / Δ236R, V234R / A 235K / Δ236R, V234R / A235L / Δ236R, V234R / A235Q / Δ236R, V234R / A235R / Δ2 36R, V234R / A235T / Δ236R, V234S / A235D / Δ236R, V234S / A235E / Δ236R, V23 4S / A235G / Δ236R, V234S / A235H / Δ236R, V234S / A235I / Δ236R, V234S / A23 5L / Δ236R, V234S / A235R / Δ236R, V234S / A235T / Δ236R, V234S / A235V / Δ236 R, V234T / Δ236R, V234T / A235I / Δ236R, V234T / A235K / Δ236R, V234T / A235Q / Δ236R, V234T / A235R / Δ236R, V234T / A235S / Δ236R, V234T / A235T / Δ236R,A protein is provided comprising a human IgG2 variant Fc region comprising a set of amino acid substitutions selected from V234T / A235V / Δ236R.

[0152] In some embodiments, V234A / Δ236R, V234A / A235T / Δ236R, V234D / A235H / Δ236R, V234D / A235K / Δ236R, V234D / A235Q / Δ236R, V234D / A235S / Δ236R, V234D / A235T / Δ236R, V234E / A235D / Δ236R, V234E / A235H / Δ236R, V234E / A235I / Δ236R, V234E / A235V / Δ236R, V234G / A235H / Δ236R, V234G / A235Q / Δ236R, V234G / A 235S / Δ236R, V234H / A235I / Δ236R, V234H / A235S / Δ236R, V234K / A235Q / Δ23 6R, V234K / A235R / Δ236R, V234K / A235S / Δ236R, V234K / A235T / Δ236R, V234K / A235V / Δ236R, V234Q / A235A / Δ236R, V234Q / A235D / Δ236R, V234Q / A235H / Δ 236R, V234Q / A235Q / Δ236R, V234Q / A235R / Δ236R, V234Q / A235S / Δ236R, V23 4Q / A235T / Δ236R, V234Q / A235V / Δ236R, V234R / A235D / Δ236R, V234R / A235 E / Δ236R, V234R / A235H / Δ236R, V234R / A235I / Δ236R, V234R / A235K / Δ236R, V234R / A235L / Δ236R, V234R / A235Q / Δ236R, V234R / A235R / Δ236R, V234R / A2 35T / Δ236R, V234S / A235D / Δ236R, V234S / A235E / Δ236R, V234S / A235G / Δ236 R, V234S / A235H / Δ236R, V234S / A235I / Δ236R, V234S / A235L / Δ236R, V234S / A235R / Δ236R, V234S / A235T / Δ236R, V234S / A235V / Δ236R, V234T / Δ236R, V2 34T / A235I / Δ236R, V234T / A235K / Δ236R, V234T / A235Q / Δ236R, V234T / A235 R / Δ236R, V234T / A235S / Δ236R, V234T / A235T / Δ236R, V234T / A235V / Δ236R,A protein is provided comprising a human IgG2 variant Fc region comprising a set of amino acid substitutions selected from A235T / Δ236R.

[0153] In some embodiments, V234A / Δ236R, V234A / A235T / Δ236R, V234D / A235H / Δ236R, V234D / A235K / Δ236R, V234D / A235Q / Δ236R, V234D / A235S / Δ236R, V234D / A235T / Δ236R, V234E / A235D / Δ236R, V234E / A235H / Δ236R, V234E / A235I / Δ236R, V234E / A235V / Δ236R, V234G / A235Q / Δ236R, V234G / A235S / Δ236R, V234Q / A235A / Δ236R, V234Q / A235D / Δ236R, V234Q / A235Q / Δ236R, V234Q / A235S / Δ236R, V234Q / A235T / Δ236R, V234Q / A235V / Δ236R, V234R / A235D / Δ236R, V234R / A235E / Δ236R, V234S / A235D / Δ236R, V234S / A235E / Δ236R, V234S / A235G / Δ236R, V234S / A235I / Δ236R, V234S / A235L / Δ236R, V234S / A235T / Δ236R, V234S / A235V / Proteins are provided that comprise a human IgG2 variant Fc region comprising a set of amino acid substitutions selected from: Δ236R, V234T / Δ236R, V234T / A235I / Δ236R, V234T / A235Q / Δ236R, V234T / A235S / Δ236R, V234T / A235T / Δ236R, V234T / A235V / Δ236R, A235T / Δ236R.

[0154] In some embodiments, V234A / A235S, V234D / A235H, V234D / A235K, V234D / A235Q, V234D / A235S, V234D / A235T, V234E / A235D, V234E / A235H, V234E / A235I, V234E / A235V, V234G / A235H, V234G / A235Q, V234G / A235S, V234H / A235I, V234H / A235S, V234K / A235Q, V234K / A235R, V234K / A235S, V234K / A235T, V234K / A235V, V234Q / A23 5A, V234Q / A235D, V234Q / A235H, V234Q / A235Q, V234Q / A235R, V234Q / A235S, V234Q / A235T, V234Q / A235V, V234R / A235D, V234R / A235E, V234R / A235H, V234R / A235I, V234R / A235K, V234R / A235L, V234R / A235Q, V23 4R / A235R, V234R / A235T, V234S / A235D, V234S / A235E, V234S / A235G, V234S / A235H, V234S / A235I, V234S / Proteins are provided that include a human IgG2 variant Fc region that includes a set of amino acid substitutions selected from A235L, V234S / A235R, V234S / A235T, V234S / A235V, V234T / A235I, V234T / A235K, V234T / A235Q, V234T / A235R, V234T / A235S, V234T / A235T, V234T / A235V.

[0155] In some embodiments, V234A / A235S, V234D / A235H, V234D / A235K, V234D / A235Q, V234D / A235S, V234D / A235T, V234E / A235H, V234E / A235I, V234E / A235V, V234G / A235H, V234G / A235Q, V234G / A235S, V23 4H / A235I, V234H / A235S, V234K / A235Q, V234K / A235S, V234K / A235T, V234K / A235V, V234Q / A235 A, V234Q / A235D, V234Q / A235H, V234Q / A235Q, V234Q / A235R, V234Q / A235S, V234Q / A235T, V234Q / A235V, V234R / A235D, V234R / A235E, V234R / A235I, V234R / A235L, V234R / A235Q, V234R / A235T, V 234S / A235D, V234S / A235E, V234S / A235G, V234S / A235H, V234S / A235I, V234S / A235L, V234S / A23 Proteins are provided that comprise a human IgG2 variant Fc region comprising a set of amino acid substitutions selected from: V234S / A235T, V234S / A235V, V234T / A235I, V234T / A235K, V234T / A235Q, V234T / A235R, V234T / A235S, V234T / A235T, V234T / A235V.

[0156] In some embodiments, V234A / A235S, V234D / A235H, V234D / A235K, V234D / A235Q, V234D / A235S, V234D / A235T, V234E / A235D, V234E / A235H, V234E / A235I, V234E / A235V, V234G / A235Q, V234G / A235S, V234H / A235I, V234H / A235S, V234K / A235Q, V234K / A235R, V234K / A235S, V234K / A235T, V234K / A235V, V234Q / A235Q, V234Q / A235R, V234Q / A235S, V234Q / A235T, V234R / A235D, V234R / A235E, V234R / A235H, V234R / A235I, V234R / A235K, V234R / A235L, V234R / A235Q, V234R / A235R, V234R / A235T, V234S / A235D, V234S / A235E, V234S / A235G, V234S / A235H, V234S / A235I, V234S / A235L, Proteins are provided that include a human IgG2 variant Fc region that includes a set of amino acid substitutions selected from V234S / A235R, V234S / A235T, V234S / A235V, V234T / A235I, V234T / A235K, V234T / A235Q, V234T / A235R, V234T / A235S, V234T / A235T.

[0157] In some embodiments, V234D / A235K, V234D / A235Q, V234D / A235S, V234D / A235T, V234E / A235I, V234K / A235Q, V234K / A235R, V234K / A235S, V234K / A235T, V234R / A235D, V234R / A235E, V234R / A235I, Proteins are provided that comprise a human IgG2 variant Fc region comprising a set of amino acid substitutions selected from V234R / A235K, V234R / A235L, V234R / A235Q, V234R / A235R, V234R / A235T, V234S / A235I, V234S / A235L, V234S / A235R, V234S / A235T.

[0158] In some embodiments, F234A / L235A / G236R, F234A / L235S / G236R, F234A / L235T / G236R, F234D / L235H / G236R, F234D / L235K / G236R, F234D / L235Q / G236R, F234D / L235S / G236R, F234D / L235T / G236R, F234E / L235D / G236R, F234E / L235H / G236R, F234E / L235I / G236R, F234E / L235V / G236R, F234G / L235H / G236R, F 234G / L235Q / G236R, F234G / L235S / G236R, F234H / L235I / G236R, F234H / L23 5S / G236R, F234K / L235Q / G236R, F234K / L235R / G236R, F234K / L235S / G236R , F234K / L235T / G236R, F234K / L235V / G236R, F234Q / L235A / G236R, F234Q / L 235D / G236R, F234Q / L235H / G236R, F234Q / L235Q / G236R, F234Q / L235R / G23 6R, F234Q / L235S / G236R, F234Q / L235T / G236R, F234Q / L235V / G236R, F234 R / L235D / G236R, F234R / L235E / G236R, F234R / L235H / G236R, F234R / L235I / G236R, F234R / L235K / G236R, F234R / G236R, F234R / L235Q / G236R, F234R / L2 35R / G236R, F234R / L235T / G236R, F234S / L235D / G236R, F234S / L235E / G236 R, F234S / L235G / G236R, F234S / L235H / G236R, F234S / L235I / G236R, F234S / G236R, F234S / L235R / G236R, F234S / L235T / G236R, F234S / L235V / G236R, F2 34T / L235A / G236R, F234T / L235I / G236R, F234T / L235K / G236R, F234T / L235 Q / G236R, F234T / L235R / G236R, F234T / L235S / G236R, F234T / L235T / G236R,A protein is provided comprising a human IgG4 variant Fc region comprising a set of amino acid substitutions selected from F234T / L235V / G236R.

[0159] In some embodiments, F234A / L235A / G236R, F234A / L235S / G236R, F234A / L235T / G236R, F234D / L235H / G236R, F234D / L235K / G236R, F234D / L235Q / G236R, F234D / L235S / G236R, F234D / L235T / G236R, F234E / L235D / G236R, F234E / L235H / G236R, F234E / L235I / G236R, F234E / L235V / G236R, F234G / L235H / G236R, F 234G / L235Q / G236R, F234G / L235S / G236R, F234H / L235I / G236R, F234H / L23 5S / G236R, F234K / L235Q / G236R, F234K / L235R / G236R, F234K / L235S / G236R , F234K / L235T / G236R, F234K / L235V / G236R, F234Q / L235A / G236R, F234Q / L 235D / G236R, F234Q / L235H / G236R, F234Q / L235Q / G236R, F234Q / L235R / G23 6R, F234Q / L235S / G236R, F234Q / L235T / G236R, F234Q / L235V / G236R, F234 R / L235D / G236R, F234R / L235E / G236R, F234R / L235H / G236R, F234R / L235I / G236R, F234R / L235K / G236R, F234R / G236R, F234R / L235Q / G236R, F234R / L2 35R / G236R, F234R / L235T / G236R, F234S / L235D / G236R, F234S / L235E / G236 R, F234S / L235G / G236R, F234S / L235H / G236R, F234S / L235I / G236R, F234S / G236R, F234S / L235R / G236R, F234S / L235T / G236R, F234S / L235V / G236R, F2 34T / L235A / G236R, F234T / L235I / G236R, F234T / L235K / G236R, F234T / L235 Q / G236R, F234T / L235R / G236R, F234T / L235S / G236R, F234T / L235T / G236R,A protein is provided comprising a human IgG4 variant Fc region comprising a set of amino acid substitutions selected from F234T / L235V / G236R, L235T / G236R.

[0160] In some embodiments, F234A / L235A / G236R, F234A / L235S / G236R, F234A / L235T / G236R, F234D / L235H / G236R, F234D / L235K / G236R, F234D / L235Q / G236R, F234D / L235S / G236R, F234D / L235T / G236R, F234E / L235D / G236R, F234E / L235H / G236R, F234E / L235I / G236R, F234E / L235V / G236R, F234G / L235Q / G236R, F234G / L235S / G23 6R, F234Q / L235A / G236R, F234Q / L235D / G236R, F234Q / L235Q / G236R, F234Q / L235S / G236R, F234Q / L235T / G236R, F234Q / L235V / G236R, F234R / L235D / G236R, F234R / L235E / G236R, F234S / L235D / G236R, F234S / L23 5E / G236R, F234S / L235G / G236R, F234S / L235I / G236R, F234S / G236R, F234S / L235T / G236R, F234S / L235V / Proteins are provided that comprise a human IgG4 variant Fc region comprising a set of amino acid substitutions selected from G236R, F234T / L235A / G236R, F234T / L235I / G236R, F234T / L235Q / G236R, F234T / L235S / G236R, F234T / L235T / G236R, F234T / L235V / G236R, L235T / G236R.

[0161] various species In some embodiments, the Fc region that is altered to create the variant Fc region may be from any mammalian species. In some embodiments, the altered Fc region is selected from mouse, rat, rabbit, rhesus monkey, or cynomolgus monkey. In some embodiments, the altered Fc region is selected from any IgG subclass. In some embodiments, the altered Fc region is selected from mouse IgG1, mouse IgG2a, mouse IgG2b, mouse IgG3, rat IgG1, rat IgG2a, rat IgG2b, or rat IgG2c.

[0162] Additional amino acid changes In some embodiments, any of the proteins comprising a variant Fc region as described above or obtained from the methods described above may also contain additional amino acid alterations, which may be insertions, deletions, or substitutions, or which may confer other desired properties, such as any one or more of: (a) altered FcRn binding activity, which may be higher or lower; (b) altered pharmacokinetic parameters, which may include slower or more rapid clearance; (c) altered immunogenicity, which may be higher or lower; (d) altered glycosylation or abolished glycosylation; (e) improved homogeneity and / or stability during manufacture and / or storage; (f) ability to form heterodimeric structures (e.g., for generating bispecific antibodies); (g) ability to form only monomeric structures; (h) ability to form multimeric structures (e.g., hexamers); (i) ability to form site-specific conjugates; (j) ability to bind to antigen; or any other desired property known in the art. It is envisioned that by combining appropriate amino acid changes, one or more of these desirable properties can be combined in a protein comprising a variant Fc region.

[0163] Amino acid residues involved in binding to FcRn include Thr250, Met252, Ile253, Ser254, Thr256, Lys288, Val305, Thr307, Val308, Leu309, His310, Gln311, Asp312, Leu314, Lys317, Lys360, Gln362, Ala378, Glu380, Glu382, Ser415, Ser424, Met428, His433, Asn434, His435, Tyr436, and Thr437 (Shields 2001; Kim 1999; Hinton 2004; Yeung 2009). Not all changes that increase the affinity of IgG for FcRn at pH 6.0 necessarily result in an increase in half-life. Examples of amino acid modifications that have been reported to increase binding to FcRn at pH 6.0 and also increase the elimination phase half-life of human FcRn in monkeys or transgenic mice include T250Q / M428L, V308P, M428L, M252Y / S254T / T256E, M428L / N434S, N434A, N434H, T307A / E380A / N434A, H433K / N434F, and L309D / Q311H / N434S (Datta-Mannan 2011; Petkova 2006; Deng 2010; Vaccaro 2005; Lee 2019). In some embodiments, any of the variant Fc regions described above comprising an amino acid change at one or more of residues 234, 235, and 236 may also comprise an amino acid substitution at one, two, or three positions selected from 250, 252, 253, 254, 256, 288, 305, 307, 308, 309, 310, 311, 312, 314, 317, 360, 362, 378, 380, 382, ​​415, 424, 428, 433, 434, 435, 436, 437 (EU numbering).In some embodiments, any of the variant Fc regions described above comprising an amino acid change at one or more of residues 234, 235, and 236 may also comprise an amino acid substitution selected from T250Q / M428L, V308P, M428L, M252Y / S254T / T256E, M428L / N434S, N434A, N434H, T307A / E380A / N434A, H433K / N434F, L309D / Q311H / N434S.

[0164] In Fc proteins, Asn297 is usually glycosylated. This glycosylation can be eliminated by substituting any other amino acid residue. Eliminating glycosylation can reduce heterogeneity of the resulting protein and may be particularly advantageous for production in cells other than mammalian cells. In some embodiments, any of the variant Fc regions described above that contain an amino acid alteration at one or more of residues 234, 235, and 236 may also contain an alteration at N297.

[0165] Human IgG4 has a heterogeneous structure and a tendency to form heterodimers with serum IgG4. This can be abolished by the amino acid substitution S228P (Angal 1993). IgG4 aggregation and Fab arm exchange under low pH conditions can be reduced by modifying Arg409 (WO2006033386; WO2008145142) or Lys370 (WO2010063785; US2017029521). IgG4 can also be stabilized by substituting Cys131 in combination with substituting any amino acid on the hinge region with Cys (WO2012022982). In some embodiments, any of the variant IgG4 Fc regions described hereinabove comprising an amino acid alteration at one or more of residues 234, 235 and 236 may also comprise one or more substitutions selected from S228P, S228P / R409K, R409K, K370X, C131X / S217C, C131X / K218C, C131X / Y222C, C131X / G223C (where X is any amino acid residue except C).

[0166] The C-terminus of an IgG may be heterogeneous due to the presence or absence of Lys447. This can be prevented by deleting the last one or two amino acid residues. In some embodiments, any of the variant Fc regions described above that contain an amino acid change at one or more of residues 234, 235, and 236 may also contain a deletion of K447, or a deletion of G446 and K447.

[0167] The physicochemical properties of antibodies and other proteins containing Fc regions can be optimized by selecting appropriate amino acid alterations (reviewed by Yang 2018). This can provide important advantages in terms of product manufacturing and stability. Introduction of additional intra- or inter-domain disulfide bonds in the CH2 and / or CH3 domains can stabilize the structure. For example, L242C / K334C, V240C / L334C, L242C / K334C, A287C / L306C, R292C / V302C, P343C / A431C, S375C / P396C, S375C / P396C / P445G / G446E / K447C, and P343C / A431C / P445G / G446E / K447C. Other stabilizing substitutions that do not introduce new disulfide bonds include Q295F / Y296A and D239E / L241M. In some embodiments, any of the variant Fc regions described above comprising amino acid changes at one or more of residues 234, 235, and 236 may also comprise one or more substitutions selected from L242C / K334C, V240C / L334C, L242C / K334C, A287C / L306C, R292C / V302C, P343C / A431C, S375C / P396C, S375C / P396C / P445G / G446E / K447C, P343C / A431C / P445G / G446E / K447C, Q295F / Y296A, D239E / L241M.

[0168] Bispecific or multispecific antibodies are of increasing interest, and numerous different methods for engineering them are known in the art (see, e.g., Brinkmann 2017). Formats containing heterodimeric Fc regions are particularly attractive because they allow for altered bispecificity and valency while preserving the expandability and druggability of the antibody (see, e.g., Moore 2019). A common approach is to create different, complementary mutations in the two different heavy chains so that homodimer assembly is inhibited but heterodimer assembly is promoted. For example, this can be done by substituting bulky residues in one chain and small residues in the other chain ("knob-into-hole"), or by substituting residues with opposite charges (D or E in one chain and K or R in the other).Exemplary mutation pairs include T366Y and Y407T (Ridgway 1996); S354C / T366W and Y349C / T366S / L368A / Y407V (Merchant 1998); F405L and K409R (Labrijn 2013); D399K / E356K and K409D / K392D (Gunasekaran 2010); E357Q / S364K and L368D / K370S (Moore 2019); S364H / F405A and Y349T / T394F (Moore 2011); D221E / P228E / L368E and D221R / P228R / K409R (Strop 2012); T350V / T366L / K392L / T394W and T350V / L351Y / F405A / Y407V (Von Kreudenstein 2013); K360E / K409W and Q347R / D399V / F405T (Choi 2013); K360E / K409W / Y349C and Q347R / D399V / F405T / S354C (Choi 2015a); K370E / K409W and E357N / D399V / F405T (Choi 2015b); K360D / D399M / Y407A and E345R / Q347R / T366V / K409V (Leaver-Fay 2016); Y349S / K370Y / T366M / K409V and E356G / E357D / S364Q / Y407A (Leaver-Fay 2016); L351D / L368E and L351K / T366K (Nardis 2017).Thus, in some embodiments, any of the variant Fc regions described above comprising amino acid changes at one or more of residues 234, 235, and 236 may also be selected from the group consisting of: T366Y and Y407T; S354C / T366W and Y349C / T366S / L368A / Y407V; F405L and K409R; D399K / E356K and K409D / K392D; E357Q / S364K and L368D / K370S; S364H / F405A and Y349T / T394F; D221E / P228E / L368E and D221R / P228R / K409R; T350V / T366L / K392L / T394W and T350V / L351Y / F405A / Y4 07V; K360E / K409W and Q347R / D399V / F405T; K360E / K409W / Y349C and Q347R / D399V / F405T / S354C; K370E / K409W and E357N / D399V / F405T; K360D / D399M / Y407A and E345R / Q347R / T366V / K409V Y349S / K370Y / T366M / K409V and E356G / E357D / S364Q / Y407A; L351D / L368E and L351K / T366K, wherein the first of the pair of substitutions is applied to one protein chain and the second of the pair is applied to the other protein chain.

[0169] The bivalent nature of IgG antibodies, i.e., increasing the affinity of binding to targets and potentially providing physiological effects, for example, through receptor dimerization, is often advantageous. However, these properties are not always desirable, and it may sometimes be preferable to have a monomeric antibody. This can be achieved by making amino acid changes in the Fc region that reduce the tendency of the chains to associate. In some cases, such changes may be the same as half of the pair of substitutions described above in the context of bispecific antibodies, since the loss of the tendency to form homodimers is a characteristic of such changes. Many other amino acid substitutions that result in the formation of a monomeric Fc region have been described, such as T394D, F408R (Wilkinson 2013); L351F / T366R / P395K / F405R / Y407E (Shan 2016); F405Q (Rose 2011); L351Y / T366Y / L368A / P395R / F405R / Y407M / K409A;L351S / T366R / L368H / P395K / F405E / Y407K / K409A; Examples include L351K / T366S / P395K / F405R / Y407A / K409Y (Ying 2012); L351S / T366R / L368H / P395K (Ying 2014); S364N / Y407N / K409T, F405N / Y407T (Ishino 2013). Thus, in some embodiments, any of the variant Fc regions described hereinabove comprising amino acid changes at one or more of residues 234, 235 and 236 also include: T394D, F408R, L351F / T366R / P395K / F405R / Y407E; F405Q, L351Y / T366Y / L368A / P395R / F405R / Y407M / K409A; L351S / T366R / L368H / P395K / F405E / Y407K / K409A; It may contain substitutions selected from L351K / T366S / P395K / F405R / Y407A / K409Y, L351S / T366R / L368H / P395K, S364N / Y407N / K409T, F405N / Y407T.

[0170] IgG antibodies can assemble into well-ordered hexamers on the cell surface after binding to their antigen.

[0171] These hexamers bind to the first component of complement C1 and induce CDC. Mutations that enhance hexamer formation and complement activation have been identified (de Jong 2016). Examples include substitutions at E345, E430, and S440. Some substitutions, such as E345R and E430F, promoted hexamer formation in solution, whereas others promoted hexamer formation only upon antibody binding to its target. An alternative method to promote hexamer formation is to create the substitutions L309C / H310L and add a portion of the C-terminal sequence of IgM or IgA (Mekhaiel 2011). If controlled aggregation without complement activation is desired, combining one of the aforementioned Fc variants with a hexamer-promoting mutation may be desirable. Thus, in some embodiments, any of the variant Fc regions described above that include an amino acid change at one or more of residues 234, 235 and 236 may also include a substitution at E345, E430 or S440, or the substitution L309C / H310L, along with the addition of an appropriate fragment of IgA or IgM.

[0172] Antibody-drug conjugates are promising antitumor agents. However, traditional methods for attaching drugs or other functional moieties to proteins result in heterogeneous products due to variable attachment to multiple different sites on the protein. To overcome this problem, several methods for site-specific modification have been developed, for example, by engineering specific sites to contain cysteine, glutamine, unnatural amino acids, short peptide tags, or glycans (Zhou 2017). While cysteines can be inserted or substituted at various locations in the Fc region, partially solvent-accessible sites may be preferred, as conjugates to such Cys residues have shown the greatest stability in plasma. Suitable sites include T289C, A339C, and S442C. In a systematic screen of all possible conjugation sites, more than 30 sites were identified in the heavy chain constant region that provided high stability for conjugation, including the following residues: 120, 166, 172, 178, 187, 199, 203, 209, 262, 336, 337, 344, 345, 382, ​​388, 411, 421, 424, 438, and 443 (Ohri 2018). In some embodiments, any of the variant Fc regions described above that contain an amino acid change at one or more of residues 234, 235, and 236 may also contain one or more amino acid changes designed to create suitable sites for site-specific conjugation of a drug or other functional moiety, such as, for example, one or more at sites selected from 120, 166, 172, 178, 187, 199, 203, 209, 262, 289, 336, 337, 339, 344, 345, 382, ​​388, 411, 421, 424, 438, 442, and 443.

[0173] The immunoglobulin constant domains within the Fc region fold in a manner highly homologous to the variable domains. Such constant domains, particularly the C HThe three domains can provide a scaffold onto which amino acid changes can be made to create new antigen-binding sites (Wozniak-Knopp 2017). Such constructs may be particularly useful in the construction of bispecific antibodies. For example, residues 358-362, 413-415, and 418-422 (which together form a continuous, solvent-exposed surface) are candidates for mutation to create new antigen-binding sites. Other residues may also be mutated to improve binding characteristics, including, for example, residues 355-357, 383-389, insertion of up to five amino acids between 389 and 390, 416, 417, and 440-447 (Wozniak-Knopp 2017). Thus, in some embodiments, any of the previously described variant Fc regions containing amino acid changes at one or more of residues 234, 235, and 236 may also contain one or more amino acid changes designed to create antigen-binding sites. In some embodiments, any of the variant Fc regions described above that contain amino acid changes at one or more of residues 234, 235, and 236 may also contain one or more amino acid changes at any number of residues selected from 355-362, 383-389, 413-422, and 440-447 to create an antigen-binding site.

[0174] Antibodies, antibody mimetics and their fragments In some embodiments, any of the proteins comprising a variant Fc region as described above or obtained from the methods described above also comprises an antigen-binding domain. The antigen-binding domain can be a full-length antibody, such as a chimeric antibody, a humanized antibody, a human antibody, a non-human antibody, or a fragment thereof. The antigen-binding domain may comprise a combination of heavy and light chain variable regions, or may comprise a single-chain variable region (scFv), or may comprise a single variable region. The antigen-binding domain may be a diabody, a Fab fragment, or a F(ab)2 fragment. The protein comprising a variant Fc region may itself be a full-length antibody, such as a chimeric antibody, a humanized antibody, or a human antibody. Such a protein may be a bispecific antibody, a multispecific antibody, or any other type of antibody known in the art, provided that it comprises an Fc region. Antigen-binding domains may also be derived from non-immunoglobulin scaffolds (e.g., adnectins, affibodies®, affilin®, affimer®, alphabodies®, anticalins®, avimers, darpins®, fynomers®, glubodies, knottins, Kunitz domains, monobodies®, nanoclamps, tetranectins (see, e.g., Simeon 2018).

[0175] Virtually any molecule can be targeted by an antigen-binding domain, and any antigen-binding domain known in the art or yet to be discovered can be incorporated into the proteins of the invention.

[0176] fusion proteins In some embodiments, any of the proteins comprising a variant Fc region as described above or obtained from the methods described above is a fusion protein. Such fusion proteins may contain one or more additional domains with any desired properties. For example, the additional domain may be a binding protein, receptor, enzyme, or cytokine. The additional domain may be derived from any source, e.g., prokaryote, eukaryote, plant, animal, mammal, human, or synthetically derived. Many such fusion proteins are known in the art, including, for example, etanercept, alefacept, abatacept, rilonacept, romiplostim, belatacept, and aflibercept (Beck 2011; Jafari 2017). In some embodiments, the variant Fc region is fused to a polypeptide for which it is desirable to increase the size, solubility, expression yield, and / or serum half-life. In some embodiments, the variant Fc region is fused to a polypeptide as a tag for polypeptide purification and / or detection.

[0177] Conjugates In some embodiments, any of the proteins comprising a variant Fc region as described above or obtained from the methods described above may be conjugated proteins. Such conjugated proteins may contain one or more additional moieties having any desired properties. For example, the additional moiety may be a binding protein, receptor, enzyme, cytokine, toxin, drug, hapten, label, chelator, radioisotope, affinity tag, linker, peptide, nucleic acid, or carbohydrate. Such additional moieties may be attached to the protein comprising a variant Fc region by any method known in the art.

[0178] Methods for producing proteins containing variant Fc regions In some embodiments, any of the proteins comprising a variant Fc region as described above or obtained from the methods described above may be produced by chemical synthesis, or preferably by recombinant expression techniques.

[0179] Monoclonal antibodies can be produced using a variety of techniques that are conventional and well known in the art, including, for example, (a) immunization of animals (e.g., mice, rats, rabbits, sheep, camels, and sharks, including animals that may be transgenic for human immunoglobulin genes) followed by hybridoma production and selection or isolation of antigen-specific B cells; (b) isolation of antigen-specific B cells from non-immunized animals, such as humans; and (c) display technologies, such as phage, yeast, ribosomal, or mammalian display. Newer techniques for monoclonal antibody production are also contemplated, including, for example, immunization of animals followed by sequencing of the resulting polyclonal antibodies and production of the corresponding monoclonal antibodies by recombinant DNA technology. Antibody mimetics can be produced using a variety of display technologies that utilize non-immunoglobulin scaffolds (see, for example, Simeon 2018).

[0180] Recombinant proteins can be produced using a wide variety of techniques that are conventional and well known in the art. A typical process involves the following steps: a) Designing the amino acid sequence of the recombinant protein. If it is to contain an antigen-binding domain from an antibody, antibody fragment, or antibody mimetic, the sequence of the antigen-binding domain is determined either by sequencing the protein or by sequencing the DNA or RNA encoding it. If the recombinant protein is to contain one or more domains from natural sources, the sequences of such domains may be obtained from an appropriate database. Protein domains may be connected by suitable peptide linkers, typically short, flexible strings of amino acids. Desired amino acid changes may be introduced into the designed sequence.

[0181] b) Designing DNA encoding the desired amino acid sequence. Due to redundancy in the genetic code, many possible nucleic acid sequences can encode the same protein. It is desirable to optimize the properties of the nucleic acid sequence for convenient subsequent manipulations and to ensure high levels of protein expression. Such optimizations can include removing or adding restriction enzyme sites, codon optimization to maximize the use of any codon bias in the host cell's translational machinery, removing unnecessary RNA splice sites, removing sequences that may create unwanted secondary structures in mRNA, and any other optimization techniques known in the art.

[0182] c) synthesizing DNA and inserting it into an expression vector. A gene encoding the desired amino acid sequence may be synthesized de novo by any technique known in the art, or may be created by site-directed mutagenesis from a similar existing gene. Such a gene may be inserted into a bacterial plasmid for amplification to provide a stock of DNA from which the gene may be excised and inserted into an expression vector, which may be, for example, a plasmid or a retrovirus.

[0183] d) Transfecting and expressing the protein. The expression vector is transfected or transduced into a suitable host cell, which may be, for example, a mammalian cell, such as NS0, CHO, PER.C6, or HEK. Alternatively, the host cell may be a bacterial, yeast, plant, or insect cell, or any other cell type suitable for producing recombinant proteins. The cells are cultured under suitable conditions to allow the transduced gene to be expressed, the protein to be synthesized, and (preferably) secreted from the cell. Depending on the design of the expression vector, the method for transfection or transduction, as well as the selection of transfected or transduced cells, the procedures for gene expression and protein synthesis may be transient or stable, and may be constitutive or inducible.

[0184] e) Purifying the protein. The protein is purified from the cell lysate or (preferably) the culture supernatant by any suitable technique known in the art, such as chromatography (e.g., ion exchange, affinity, and size exclusion chromatography), centrifugation, differential solubility, or any other standard technique for protein purification. Affinity purification using Protein A is a preferred method for purifying proteins containing an Fc region.

[0185] An alternative process for the production of recombinant proteins may be to use cell-free expression systems (see, e.g., Gregorio 2019). It is envisioned that the proteins of this invention may be made by such processes.

[0186] nucleic acid In some embodiments, nucleic acids are provided that encode proteins comprising a variant Fc region comprising amino acid substitutions at residues 234 and / or 235 and an Arg amino acid substitution or insertion at residue 236. The encoded protein may also have any of the functional properties and / or additional amino acid alterations described above, or any combination thereof. The nucleic acid may be DNA or RNA. It may be obtained by any suitable method known in the art, and its nucleotide sequence may be determined by any suitable method known in the art.

[0187] vector In some embodiments, a vector is provided that contains a nucleic acid encoding a protein comprising a variant Fc region comprising an amino acid substitution at residues 234 and / or 235 and an amino acid substitution or insertion of Arg at residue 236. The encoded protein may also have any of the functional properties and / or additional amino acid alterations described above, or any combination thereof. In some embodiments, the vector is capable of causing expression of the protein when introduced into a suitable host cell. In some embodiments, the vector is a self-replicating extrachromosomal vector. In some embodiments, the vector is capable of integrating into the host cell genome. Expression vectors are constructed to be compatible with the host cell type. Thus, expression vectors are utilized in the present invention, including, but not limited to, those that enable expression of proteins comprising a variant Fc region in bacteria, yeast, plant, insect, or mammalian cells, as well as in in vivo systems, such as by generating transgenic animals or plants. As is known in the art, a variety of expression vectors are commercially available or otherwise that can be used to express the proteins of the present invention. In some embodiments, the vector is a virus. In some embodiments, the vector is a virus capable of infecting human cells in vivo and causing expression of a protein of the invention. In some embodiments, the vector is an oncolytic virus. As is well known in the art, any expression vector can be associated with any suitable promoter, enhancer, or other expression-facilitating elements.

[0188] cell In some aspects, cells are provided that are capable of producing a protein comprising a variant Fc region comprising an amino acid substitution at residues 234 and / or 235 and an Arg amino acid substitution or insertion at residue 236. Examples of suitable host cells include bacteria, yeast, plant, or insect cells, or mammalian cells such as NS0, CHO, PER.C6, or HEK. Alternatively, the host cell may be any other cell type suitable for producing recombinant proteins. In some embodiments, the host cell comprises a nucleic acid encoding the protein comprising the variant Fc region that is stably integrated into the cellular genome. In other embodiments, the host cell comprises a non-integrated nucleic acid, such as a plasmid, cosmid, phagemid, or linear expression element, encoding the protein comprising the variant Fc region.

[0189] Pharmaceutical Composition In some aspects, the present invention provides pharmaceutical compositions comprising proteins comprising a variant Fc region defined in any of the aspects and embodiments herein. Pharmaceutical compositions may be formulated using pharmaceutically acceptable carriers, diluents, adjuvants, and / or excipients according to conventional techniques known in the art (see, e.g., Remington 1995; Shire 2009). Pharmaceutically acceptable carriers, diluents, adjuvants, and excipients should be suitable for the proteins of the invention and the chosen mode of administration. Suitability is determined based on the lack of significant negative impact on desired biological properties (e.g., binding activity to either antigen or receptor, stability, pharmacokinetic parameters) and the lack of toxicity to the subject to whom the composition will be administered. The amount of active ingredient (including the protein of the invention) in a pharmaceutical composition can be varied to obtain an amount effective to achieve the desired therapeutic response. The selected amount can be determined by the pharmacokinetic properties of the active ingredient, the route and timing of administration, the age, sex, weight, and medical condition of the patient being treated, and similar factors well known in the art. The pharmaceutical compositions may be administered by any parenteral route (eg, by injection or infusion).

[0190] Therapeutic uses In some aspects, the present invention provides a protein, e.g., an antibody, or pharmaceutical composition defined in any aspect or embodiment described herein for use as a medicament. In another aspect, the present invention provides a protein, e.g., an antibody, or pharmaceutical composition defined in any aspect or embodiment described herein for use in treating a disease. In another aspect, a method of treating a subject having a disease is provided, comprising administering to the subject an effective amount of a protein, e.g., an antibody, or pharmaceutical composition defined in any aspect or embodiment described herein. The antibodies or proteins of the present invention are not limited to any particular details or modalities, and the range of diseases that can be treated is extremely broad. However, a common factor in selecting a suitable disease indication is that binding to Fcγ receptors and potentially triggering inflammatory responses or engaging other effector functions that could cause adverse effects would be expected to be undesirable for a medicament. Such diseases include, for example, angioedema, arthritis, asthma, atopic dermatitis, autoimmune diseases, cancer, Castleman's disease, coagulation disorders, Crohn's disease, cryopyrin-associated periodic syndrome, diabetes, drug toxicity, eczema, graft-versus-host disease, hypercholesterolemia, hypophosphatemia, infectious diseases, ischemic heart disease, macular degeneration, multiple sclerosis, muscle wasting, osteoporosis, paroxysmal nocturnal hemoglobinuria, primary hemophagocytic lymphohistiocytosis, psoriasis, sepsis, sickle cell disease, stroke, systemic lupus erythematosus, thrombocytopenic purpura, transplantation, ulcerative colitis, or other disease indications in which therapeutic monoclonal antibodies or Fc fusion proteins have been used or tested or are currently being used or tested in human subjects. In some circumstances, the reduction in undesired or adverse reactions obtained with the use of a protein of the invention (compared to a similar protein without the amino acid changes of the invention) permits the use of higher, more frequent, or greater doses of a protein of the invention than would otherwise be expected to be considered safe or prudent for a similar protein without the amino acid changes of the invention.Thus, the proteins of the invention are predicted to have a higher therapeutic index than similar proteins without the amino acid alterations of the invention.

[0191] In some aspects, the present invention provides proteins, e.g., antibodies, or compositions defined in any aspect or embodiment described herein, that have reduced immunogenicity compared to a similar protein without the amino acid alterations of the invention. Unwanted immunogenicity of biological drugs, particularly proteins, is problematic because it leads to the formation of anti-drug antibodies that can neutralize the therapeutic activity of the drug or cause adverse reactions or side effects. Such anti-drug antibodies limit the duration of effective treatment. Binding to Fc receptors can enhance the immunogenicity of proteins containing Fc regions (Chen 2007, Loureiro 2011). Proteins of the invention with reduced immunogenicity are particularly useful for therapeutic applications for a number of reasons, including, for example, greater efficacy, lower toxicity, fewer side effects, and longer duration of effective therapeutic activity.

[0192] Research and diagnostic uses In some aspects, the present invention provides a protein, e.g., an antibody, or composition defined in any aspect or embodiment described herein for use in research, diagnostic, or quality control testing. Accordingly, the present invention provides research and diagnostic methods and quality control methods and compositions using the proteins described herein. Such methods and compositions can be used to detect or identify disease, monitor treatment progress, assess post-treatment status, monitor recurrence after treatment, assess risk of developing disease, test for activity of biological drugs in vitro, and the like. In some aspects, the diagnostic method or composition is used ex vivo, e.g., to detect or measure the level of an antigen recognized by an antibody or protein of the invention. In other aspects, the diagnostic method or composition is used in vivo, e.g., to determine internalization of an antigen recognized by an antibody or protein of the invention. Because the antibodies or proteins of the invention are not limited to any particular details or modalities, the scope of research, quality control, and diagnostic testing is extremely broad and may encompass almost any antigenic target, known or yet to be discovered, and may utilize any relevant technology known in the art, such as enzyme-linked immunosorbent assays, cell-based assays, radioimmunoassays, fluorescence assays, flow cytometry, surface plasmon resonance, chemiluminescence, electrochemiluminescence, chromatographic assays, homogeneous immunoassays (e.g., AlphaLISA®, Lumit®, CEDIA®), immunohistological assays, Western blots, immunoprecipitation, lateral flow tests, etc. In some embodiments, the antibodies or proteins of the invention are conjugated proteins containing moieties suitable for detection in a selected test system. In some embodiments, the antibodies or proteins of the invention are radiolabeled for use in in vivo diagnostic imaging tests.However, a common factor in the selection and design of a test or composition is the desire to avoid unwanted binding of the antibodies or proteins of the invention to Fcγ receptors or C1q, or engaging in other effector functions that could otherwise interfere with the assay or cause false-positive or false-negative responses. Research, quality control, and diagnostic uses for the proteins of the invention include, for example, any type of test that measures binding to cells or protein arrays that can display Fcγ receptors. Binding to such receptors can result in a "false-positive" response. Such unwanted binding can be reduced or eliminated by use of the proteins of the invention. Other testing uses for the proteins of the invention include, for example, any type of test in which the proteins of the invention are used as a control sample due to their lack of binding to Fcγ receptors or C1q.

[0193] In some embodiments of the present invention, there is provided a method for reducing unwanted binding to Fcγ receptors of an antibody or Fc fusion protein or antibody conjugate or Fc fusion protein conjugate for use in a research or diagnostic test system, comprising the steps of (a) producing a variant form of said antibody or Fc fusion protein or conjugate containing an amino acid alteration according to the invention, and (b) substituting said variant form for the original antibody or Fc fusion protein or conjugate. The test system may be, for example, an enzyme-linked immunosorbent assay (ELISA), a fluorescence-based assay, an immunohistochemistry test, a chemiluminescence or electrochemiluminescence test, a diagnostic imaging test, or any other immunological test system. [Example]

[0194] The following examples are provided to illustrate the practice of this invention. They are not intended to limit or define the entire scope of this invention. Due to differences in immunoglobulin alleles or due to different cloning strategies, there may be slight differences (i.e., amino acid changes) between the amino acid sequences described herein and those reported elsewhere in the literature. Such differences do not limit the scope of this invention.

[0195] Example 1 Production of antibodies containing variant Fc regions 1.1 Design, Expression and Purification Codon-optimized synthetic genes encoding the wild-type human IgG1 heavy chain constant region (SEQ ID NO: 4) or its variants were provided by Genewiz® in the pUC57-Kan vector. Purified DNA was digested with NheI / NotI restriction enzymes, and the genes were extracted for ligation into the pUV mammalian expression vector. Similarly, a codon-optimized synthetic gene for the human kappa light chain constant region (SEQ ID NO: 5) was ligated into the pUV vector. Synthetic genes encoding the VH and VL variable regions of anti-CD20 rituximab (SEQ ID NO: 6 and SEQ ID NO: 7), anti-CD3 muromonab (SEQ ID NO: 8 and SEQ ID NO: 9), or anti-CD52 alemtuzumab (SEQ ID NO: 10 and SEQ ID NO: 11) were provided by Genewiz® with NheI and AvaI restriction sites at the 5' and 3' ends for cloning. The VH genes were digested, excised, and ligated into the wild-type human IgG1 heavy chain pUV vector. The VL gene was digested, excised, and ligated into a human kappa light chain pUV vector. To create variants, synthetic genes containing variant Fc regions were synthesized with KasI and SacII restriction sites at the 5' and 3' ends, digested, excised, and ligated into an existing vector encoding the anti-CD20 immunoglobulin heavy chain (SEQ ID NO: 12). Variants of the anti-CD3 heavy chain (SEQ ID NO: 14) or anti-CD52 heavy chain (SEQ ID NO: 16) were generated by digesting the appropriate VH variable region with NheI and AvaI, followed by excision and ligation into an anti-CD20 variant heavy chain vector, also digested with NheI and AvaI, from which the anti-CD20 (canti-CD20) VH sequence had been removed for replacement with anti-CD3 or anti-CD52. A similar process was used to create variants of wild-type human IgG2 (SEQ ID NO: 18), human IgG4 (SEQ ID NO: 19), mouse IgG2a (SEQ ID NO: 20), rat IgG2b (SEQ ID NO: 22) or rabbit IgG (SEQ ID NO: 24). In the case of human IgG4, in addition to the amino acid changes made to alter binding to Fc receptors, an additional amino acid substitution, S228P, was introduced to stabilize the antibody structure (Angal 1993).For expression of mouse, rat, or rabbit antibodies, the anti-CD20 VL gene was ligated with a gene encoding the relevant kappa light chain constant region (SEQ ID NO: 21, SEQ ID NO: 23, or SEQ ID NO: 25). Figure 1 illustrates the amino acid sequence of the wild-type IgG Fc region used herein and the sites of amino acid changes. The sequence of the final insert was confirmed by Sanger sequencing.

[0196] Intact antibodies were produced by transient transfection of HEK293 cells with a mixture of the appropriate heavy chain expression vector and the corresponding light chain expression vector encoding the immunoglobulin light chain of anti-CD20 (rituximab) (SEQ ID NO: 13), anti-CD3 (muromonab) (SEQ ID NO: 15), or anti-CD52 (alemtuzumab) (SEQ ID NO: 17). HEK293 cells were grown in suspension in serum-free culture medium (ThermoFisher catalog no. 12338026) and transfected using linear polyethyleneimine at a concentration of 2 x 10 viable cells. 6 Cells were transfected at a cell count of 1000 / mL. Cells were cultured in Erlenmeyer flasks at 37°C with shaking at 140 rpm in 5% CO2 for 6 days. Fc fusion proteins were produced in a similar manner by transient transfection of HEK293 cells with expression vectors encoding the required fusion proteins. Alternatively, antibodies were produced by transient transfection of CHO cells using a similar mixture of heavy and light chain expression vectors. Suspension-adapted CHO cells were cultured in serum-free medium and 2 x 10 cells were added to MaxCyte buffer. 8 The cells were resuspended at 1000 cells / mL and transfected by electroporation using a MaxCyte STC instrument (MaxCyte, Gaithersburg, USA) according to the manufacturer's instructions. Cells were incubated for 30 min and then cultured at 37°C for 24 h and 32°C for an additional 10 days. Cells were harvested by centrifugation at 3000g for 45 min, and the supernatant was clarified using a 0.22 μm vacuum-driven filter.

[0197] Culture supernatants were analyzed by non-reducing SDS-polyacrylamide gel electrophoresis (SDS-PAGE) followed by staining with Coomassie blue. This showed that all samples yielded a single band corresponding to a molecular weight of approximately 150 kDa. The antibody was purified using a 1 mL MabSelect SuRe® Protein A chromatography column (GE Life Sciences), eluted with 0.1 M sodium citrate buffer, pH 3.0, and then neutralized with 10% (v / v) Tris base, pH 9.0. The eluted antibody was buffer exchanged with PBS, pH 7.2, and concentrated using a centrifugal concentrator as needed. Protein concentration was measured by absorbance at 280 nm using a Nanodrop® ND-1000 spectrophotometer (Thermo-Fisher) and adjusted to 1 mg / mL by dilution with PBS. Several separate transfection experiments were performed. Table 5 lists the samples prepared.

[0198] [Table 6-1]

[0199] [Table 6-2]

[0200] [Table 6-3]

[0201] [Table 6-4]

[0202] [Table 6-5]

[0203] [Table 6-6]

[0204] Table 6-7

[0205] 1.2 Antigen binding The antigen-binding properties of representative CD3 and CD52 antibodies were tested by enzyme-linked immunosorbent assay (ELISA). All steps were performed at room temperature (18–22°C). 50 μL of either (a) 1 μg / mL CD3D / CD3E heterodimer (Sino Biologicals, catalog no. CT038-H2508H) in PBS or (b) 5 μg / mL sheep Fc-CD52 fusion protein (Absolute Antibody, catalog no. PR00205) in PBS was added to a 96-well microplate (Sigma, catalog no. M9410-ICS) and incubated with shaking for approximately 2 hours. The coating antigen was removed, and 200 μL of PBS containing 1% (w / v) casein and 0.05% (v / v) Proclin (blocking buffer) was added and incubated for approximately 3 hours. The plate was rinsed with PBS containing 0.05% (v / v) Tween 20 (wash buffer), and test samples diluted in blocking buffer to a final concentration of 5 μg / mL were added. The plate was incubated with shaking for approximately 1 hour and then rinsed with washing buffer. HRP-conjugated rabbit anti-human IgG (Sigma catalog no. A8972) was diluted 1:10,000 in blocking buffer, and 100 μL was added to each well. The plate was incubated with shaking for approximately 2.5 hours, then washed four times with washing buffer and twice with water. 100 μL of 3,3',5,5'-tetramethylbenzidine liquid substrate, ultrasensitive (TMB) (Abcam catalog no. ab171523) was added, and the plate was incubated for approximately 10 minutes. The reaction was stopped by adding 50 μL of 1 M sulfuric acid, and the absorbance was read at 450 nm minus 620 nm using a microplate spectrophotometer (Anthos Labtec HT). Each sample was tested in quadruplicate. The mean was calculated and the mean response of the negative control (buffer alone) was subtracted. The results are shown in Table 6. All of the CD3 variant antibodies resulted in positive binding to the CD3 antigen, and all of the CD52 variant antibodies resulted in positive binding to the CD52 antigen. As specificity controls, there was no significant binding of the wild-type CD3 antibody to CD52 or the wild-type CD52 antibody to CD3.

[0206] [Table 7]

[0207] Example 2 Fcγ receptor binding 2.1 Screening a panel of variants for binding to human FcγRI Binding analysis was performed by surface plasmon resonance at 25°C using a Biacore® T200 instrument equipped with Biacore® T200 control software. The running buffer (HBS-EP+) consisted of 0.01 M HEPES pH 7.4 containing 0.15 M NaCl, 3 mM EDTA, and 0.05% v / v surfactant P20. The sample compartment was kept at 4°C during analysis. The method was run using the manufacturer's standard method wizard as described in the Biacore® T200 instrument handbook.

[0208] Polyclonal anti-histidine antibody (GE Healthcare catalog number 28-9950-56) was immobilized on a CM5 chip using EDC / NHS chemistry. Cells were activated by treatment with EDC / NHS for 7 minutes. The antibody was diluted to 5 μg / mL in 10 mM sodium acetate, pH 4.5, and injected at 5 μL / min for 7 minutes. The flow cell was deactivated by injection of 1 M ethanolamine for 7 minutes. This process was repeated on a separate flow cell in the absence of anti-his antibody to generate a reference surface. Recombinant human FcγRI (CD64) with a histidine tag (Sino Biologicals catalog number 12056-H08H) was diluted to 10 μg / mL in HBS-EP and injected at a flow rate of 30 μL / min for 2 minutes. Samples of undiluted culture supernatant from HEK293 cells expressing a human IgG1 antibody containing a variant Fc region were injected at 30 μL / min for 1 minute and allowed to dissociate for 2 minutes. All test samples showed substantially reduced binding to FcγRI compared to the control wild-type IgG antibody and completely dissociated within 2 minutes. Therefore, regeneration between cycles was not necessary. Responses to the test flow cell (coupled with anti-His antibody) were analyzed after subtracting the response to the control flow cell (mock-coupled). Relative binding responses were measured 55 seconds after the start of sample injection. To account for any possible drift in response due to slow dissociation of the His-tagged FcγRI, binding responses were measured against a baseline interpolated between reporting points placed before sample injection and at the end of dissociation.

[0209] The entire experiment was repeated three times. Due to the relatively crude nature of the samples (variable refractive index, presence of culture medium and other cellular products), some sensorgrams were insufficient and were excluded from analysis. The mean and standard deviation of the binding response were then calculated for each sample. The average of all standard deviations was 6.7 RU. The highest mean response of 65.0 RU was exhibited by sample 1-167 (L234A / L235A). All variants containing the G236R mutation exhibited lower mean responses. Sample 1-119 (L234Q / L235S / G236R) was taken as a representative variant that was effectively silenced by the initial experiment. It exhibited a mean response of 11.9 RU. A cutoff of 21.9 RU was calculated to be equal to the mean response of this representative variant plus 1.645 times the mean standard deviation of the samples. Samples yielding responses above this cutoff were considered to yield significantly greater binding to FcγRI compared to representative sample 1-119.

[0210] [Table 8-1]

[0211] [Table 8-2]

[0212] [Table 8-3]

[0213] [Table 8-4]

[0214] [Table 8-5]

[0215] This experiment identified a number of variants that conferred low levels of binding to human FcγRI, but many others conferred binding greater than a cutoff equal to the mean of a representative low-binding sample (1-119) plus 1.645 times the standard deviation of the response. Variants that conferred these higher levels of binding were excluded from subsequent analysis.

[0216] 2.2 Screening of a panel of purified variants for binding to human FcγRI Based on the results of the culture supernatant screening, variants that conferred low levels of FcγRI binding were selected for further analysis as purified antibodies. Binding analysis was performed by surface plasmon resonance at 25°C using a Biacore® T200 instrument equipped with Biacore® T200 control software. The sample compartment was maintained at 5°C. The running buffer (HBS-EP+) consisted of 0.01 M HEPES pH 7.4 containing 0.15 M NaCl, 3 mM EDTA, and 0.05% v / v surfactant P20. The method was run using the manufacturer's standard method wizard as described in the Biacore® T200 instrument handbook.

[0217] A mouse monoclonal anti-histidine antibody (Qiagen catalog no. 34670) was immobilized on a CM5 chip using EDC / NHS chemistry. Cells were activated by treatment with EDC / NHS for 7 minutes. The antibody was diluted to 10 μg / mL in 10 mM sodium acetate, pH 5.0, and injected at 5 μL / min for 7 minutes. The flow cell was deactivated by injection of 1 M ethanolamine for 7 minutes. This process was repeated on a separate flow cell in the absence of anti-his antibody to generate a reference surface. Recombinant human FcγRI (CD64) with a histidine tag (Sino Biological catalog no. 12056-H08H) was diluted to 5 μg / mL in HBS-EP and injected at a flow rate of 10 μL / min for 2 minutes. A sample of purified antibody containing a variant Fc region (1 mg / mL in PBS) was diluted to 100 μg / mL in running buffer and injected at 30 μL / min for 1 minute, followed by a 2-minute dissociation period. Between each cycle, the chip was regenerated by a 0.5-minute injection of 10 mM glycine, pH 3.0, at a flow rate of 30 μL / min, and then reloaded with a new injection of FcγRI. At the beginning of each experiment, three blank cycles were performed in which buffer alone was injected instead of the test sample. The average SPR signal (relative to the reference cell) was measured approximately 8-13 seconds before the injection of the test sample (baseline) and approximately 3-8 seconds before the end of the injection (response). To allow for complete dissociation of bound FcγRI and obtain a corrected response for each test sample, the response of the third blank (buffer) sample was subtracted from the response of each test sample. The entire experiment was repeated three times, once for each test sample being analyzed in a different order. Two independent lots of the LALAPG reference antibody were included in each experiment. One lot (samples 2-66) was prepared in parallel with all of the other variants and used for statistical comparison, while the other lot (sample 10.75) was prepared separately. This lot was run at the beginning and end of each experiment as a consistency check.

[0218] The mean corrected response for each sample across three experiments was compared to the mean corrected response for the LALAPG reference antibody (sample 66) using a two-tailed t-test assuming unequal variances. Calculations were performed using the T.TEST function in Microsoft Excel® 2016. Results were marked as (a) significantly lower (p≦0.05), (b) equivalent (p>0.05), or (c) significantly higher (p≦0.05) than the mean corrected response of the LALAPG reference antibody (sample 2-66). Results are shown in Table 8. The majority of the new variants (including samples 2-2 through 2-64, respectively) yielded significantly lower binding to FcγRI compared to the LALAPG reference antibody, many of which were indistinguishable from the buffer control. In contrast, all of the previously published variants (samples 2-65, 2-67, and 2-78) and controls (samples 2-79 through 2-83) yielded higher or indistinguishable binding compared to the LALAPG reference antibody. A previously published variant, sample 2-72, contained mutations designed to enhance binding to FcγRIII, resulting in binding to FcγRI comparable to the wild-type reference.

[0219] [Table 9-1]

[0220] [Table 9-2]

[0221] [Table 9-3]

[0222] This experiment shows that many variants containing amino acid substitutions at positions 234 and / or 235 along with the substitution G236R bind significantly less (p≦0.05) to human FcγRI than the LALAPG reference antibody. This result is particularly surprising considering that the LALAPG reference antibody was previously described as having "completely abolished FcγR interactions" (Schlothauer 2016). In stark contrast, none of the previously published variants (samples 2-65 and 2-67 through 2-83) showed lower binding than the LALAPG reference antibody; many of them even showed significantly higher binding (p≦0.05).

[0223] Additional variants were tested in the same manner, including three produced in CHO cells, and the results are shown in Table 9.

[0224] [Table 10]

[0225] 2.3 Binding to Mouse, Rat, and Rabbit FcγRI Compared to Human FcγRI A variant antibody with amino acid changes L234Q / L235S / G236R, representing a new variant that resulted in very low binding to human FcγRI, was compared to a control set for binding to mouse, rat, and rabbit FcγRI. Binding analysis was performed as described in Example 2.2, except that either human FcγRI, mouse FcγRI (Sino Biological Company, Catalog No. 50086-M08H), rat FcγRI (Sino Biological Company, Catalog No. 80016-R08H), or rabbit FcγRI (Sino Biological Company, Catalog No. 65010-T08H) was loaded onto the chip at 5 μg / mL each. As before, binding responses were corrected by subtracting the response obtained with a blank sample consisting of running buffer alone. The results are shown in Table 10. As previously mentioned, the L234Q / L235S / G236R variant produced a substantially lower response than the LALAPG reference for binding to human FcγRI (2.0 RU compared to 72.9 RU). It also produced substantially lower binding to rabbit FcγRI (0.9 RU compared to 5.6 RU). Neither sample showed measurable binding to mouse or rat FcγRI.

[0226] [Table 11]

[0227] 2.4 Binding to human FcγRII and human FcγRIII A variant antibody with amino acid changes L234Q / L235S / G236R, representing a new variant that conferred very low binding to human FcγRI, was compared to a control set for binding to various forms of human FcγRII and human FcγRIII. Binding analysis was performed as described in Example 2.2, except that either human FcγRIIA (R131, Sino Biological Company Catalog No. 10374-H08H), human FcγRIIB (Sino Biological Company Catalog No. 10259-H08H), human FcγRIIIA (F158, Sino Biological Company Catalog No. 10389-H08H), or human FcγRIIIB (NA2 allotype, Sino Biological Company Catalog No. 11046-H08H) was loaded onto the chip. As before, binding responses were corrected by subtracting the response obtained with a blank sample consisting of running buffer alone. The results are shown in Table 11. In this experiment, neither the L234Q / L235S / G236R variant nor the LALAPG reference antibody showed measurable binding to any of the human FcγRs.

[0228] [Table 12]

[0229] Further experiments were performed in a similar manner to measure the binding of selected variants to human FcγRIIIA, and the results are shown in Table 12. Similar to the LALAPG reference antibody, the majority of the new variants showed very low binding to FcγRIIIA.

[0230] [Table 13-1]

[0231] [Table 13-2]

[0232] 2.4 Binding to additional Fcγ receptors Additional studies were performed to measure binding of CD20, CD3, and CD52 IgG1 antibodies with the following mutations: L234G / L235S / G236R, L234S / L235T / G236R, L234S / L235T / G236R, L234S / L235V / G236R, L234T / L235Q / G236R, L234T / L235T / G236R, L234A / L235A (LALA), L234A / L235A / P329G (see LALAPG), and N297Q (aglycosyl). Binding was measured by Biacore® analysis as described in Section 2.2 using the following receptors: human FcγRI, human FcγRIIa R131 allele, human FcγRIIb, human FcγRIIIa F158 allele, human FcγRIIIa V158 allele, human FcγRIIIb NA2 allele, mouse FcγRI, and rat FcγRI. As previously described, the previously published variants LALA, LALAPG, and aglycosyl all provided binding to FcγRI significantly above the negative control. The LALA variant also provided binding to both the FcγRII allele R131 and the FcγRIIIa allele significantly above the negative control. The aglycosyl variant provided substantial binding to both mouse and rat FcγRI significantly above the negative control. In contrast, binding of all of the new variants to all of the FcγRs did not significantly exceed binding of the negative control.

[0233] Further experiments were performed to measure binding of CD20 antibodies to the Fc regions from human IgG1, IgG2, and IgG4, mouse IgG2a, rat IgG2b, and rabbit IgG. For these studies, a representative variant, L234Q / L235S / L236R, was used. Wild-type human IgG2 has a deletion at residue 236. Two variants were generated, one in which the deletion was retained (L234Q / L235S) and the other in which an Arg residue was inserted (L234Q / L235S / Δ236R). The variant IgG4 antibody contained the stabilizing mutation S228P. Two IgG1 variants also contained mutations designed to increase binding to FcRn (L234Q / L235S / L236R / M252Y / S254T / T256E and L234Q / L235S / L236R / T250Q / M428L). Binding was measured by Biacore® analysis as described in section 2.2 using the following receptors: human FcγRI, human FcγRIIa R131 allele, human FcγRIIb, human FcγRIIIa F158 allele, human FcγRIIIa V158 allele, human FcγRIIIb NA2 allele, mouse FcγRI, and rat FcγRI. The IgG2 variant L234Q / L235S / Δ236R yielded weakly positive binding responses to human and mouse FcγRI. The IgG2 variant L234Q / L235S induced a stronger response to human FcγRI but was negative for mouse FcγRI. The IgG2 variants induced negative binding responses to all other Fcγ receptors tested. The rat IgG2b variant induced positive responses to human FcγRI, FcγRIIa R131, FcγRIIb, FcγRIIIa F158, and rat FcγRI, but was negative for human FcγRIIIa V158 and mouse FcγRI. Variants based on human IgG4, mouse IgG2a, rabbit IgG, and human IgG1 with half-life-extending mutations induced negative binding responses to all Fcγ receptors tested.

[0234] Example 3 Binding to FcRn 3.1 Binding to FcRn by surface plasmon resonance A variant antibody with amino acid changes L234Q / L235S / G236R, representing a new variant that conferred very low binding to human FcγRI, was compared with a control set for binding to human and monkey FcRn. Binding analysis was performed as described in Example 2.2, except that the chip was loaded with either human FcRn (R&D Systems, Catalog No. 8639-Fc) at 5 μg / mL or cynomolgus monkey FcRn (Sino Biological, Catalog No. CT031-C08H) at 2 μg / mL, and the running buffer (HBS-EP+), which was also used to dilute all samples and adjusted to pH 6.0 with acetic acid. Human FcRn binding was measured in duplicate. As previously described, binding responses were corrected by subtracting the response obtained with a blank sample consisting of running buffer alone. The results are shown in Table 13. All of the samples conferred substantially similar overall binding to both human and cynomolgus monkey FcRn.

[0235] [Table 14]

[0236] Further experiments were performed in a similar manner to measure the binding of selected variants to human FcRn, cynomolgus monkey FcRn, and mouse FcRn (Sino Biological Catalog No. CT029-M08H). The results are shown in Table 14. All variants produced an overall similar response compared to the wild-type reference antibody.

[0237] [Table 15]

[0238] Further experiments were performed to test a larger panel of variants for human FcRn binding. Due to the relatively large number of samples, it was necessary to correct for the effects of baseline drift and loss of binding capacity during the run, which is required for repeated regeneration of the Biacore® chip. Baseline correction was calculated by interpolating the responses of two blank samples, one at the beginning and one at the end of the run, to determine an individual baseline correction for each cycle, which was subtracted from the response of the sample run in that cycle. To correct for loss of binding capacity, a normalization factor was calculated by interpolating the responses of reference sample 10.75 (LALAPG reference) at the beginning and end of the run to determine an individual normalization factor for each cycle. The baseline-corrected response was then divided by the normalization factor to obtain the corrected and normalized response, expressed as a percentage of the interpolated response of the wild-type reference. The entire experiment was performed in triplicate. The results are shown in Table 15. With the exception of sample 2-67 (N297Q, aglycosyl), all of the variants produced normalized responses within the range of 93% to 108% of the response of the wild-type reference (sample 2-1).

[0239] [Table 16-1]

[0240] [Table 16-2]

[0241] 3.2 Binding to FcRn by NanoBiT® Immunoassay An alternative method for measuring binding to FcRn in solution uses a luminescence assay. The NanoBiT® FcRn Immunoassay (Promega) is a homogeneous (no-wash) competitive assay for measuring the interaction between human FcRn and Fc proteins, including antibodies. Human IgG labeled with LgBiT (Tracer-LgBiT) is used as the tracer. C-terminally biotinylated FcRn attached to streptavidin-SmBiT (FcRn-SAv-SmBiT) is used as the target. In the presence of a sample containing no IgG, Tracer-LgBiT binds to the FcRn-SAv-SmBiT target, resulting in a maximal luminescence signal. In the case of a sample containing IgG, unlabeled IgG is expected to compete with Tracer-LgBiT for binding to the target, resulting in a concentration-dependent decrease in the luminescence signal. Experiments were performed using the NanoBiT® FcRn Immunoassay Kit (Promega catalog no. CS53019A) according to the manufacturer's instructions. Samples were diluted in an offline plate; 25 μL of each test sample (1 mg / mL in PBS) was mixed with 2.5 μL of pH adjustment buffer and 72.5 μL of FcRn assay buffer to obtain a concentration of 250 μg / mL at pH 6.0. Four-fold dilutions of a positive control antibody (human IgG) were prepared starting at 4 mg / mL. 25 μL of tracer-LgBiT was added to each well of a white microplate (Sigma catalog no. CLS3912), followed by 25 μL of sample (in duplicate). The plate was briefly mixed on a plate shaker, and 50 μL of streptavidin-SmBiT plus hFcRn-AviTag was added to each well. The plate was incubated on a shaker at room temperature for 2.5 hours. 25 μL of FcRn NanoGlo® substrate was added, and after 5 minutes, luminescence was measured in a Veritas microplate luminometer (Turner BioSystems). The calibration curve is shown in Figure 2. The raw data were fitted to a 5-parameter logistic equation, and the reference concentrations of the test samples were calculated by interpolation with the standard curve. The reference concentrations represent the concentrations of the positive control antibody that would be expected to produce the same inhibition.The mean results, expressed as a percentage of the baseline concentration of the wild-type reference, are shown in Table 16.

[0242] [Table 17]

[0243] Consistent with earlier results (Example 3.1), the aglycosyl antibody (2-67) showed reduced binding activity. There was no detectable binding activity for 2.77 (mouse IgG2a) and 8.77 (rat IgG2b). Variants 2-63 (L234T / L235T / G236R), 11.77 (IgG2 L234Q / L235S / Δ236R), 11.144 (IgG2 L234Q / L235S), and 23.77 (rabbit IgG) showed increased binding activity. Two variants (10.145 and 10.146) containing additional mutations designed to enhance binding to FcRn did not actually exhibit substantially higher binding activity. The binding activity of the other variants was within ±30% of the wild-type reference.

[0244] Taken together, these experiments demonstrate that the new variants have minimal, if any, effect on binding to FcRn.

[0245] Example 4 Binding to C1q Human C1q (Sigma catalog no. 20476) was coupled to horseradish peroxidase (HRP) using the Lightning-Link® HRP Conjugation Kit (Innova Biosciences catalog no. 701-0003) as follows: 2 μL of Lightning-Link® Modifier was gently mixed with 20 μL of C1q at 1.15 mg / mL. The mixture was added to lyophilized Lightning-Link® HRP (sufficient for up to 40 μg of protein) and gently mixed. The mixture was incubated at 4°C for 3 hours. 2 μL of Lightning-Link® Quencher was added and incubated at 4°C for 30 minutes. The resulting HRP-labeled C1q was diluted to 20 μg / mL in phosphate-buffered saline (PBS) containing 1% casein (blocking buffer) and stored at 4°C.

[0246] Binding analysis was performed by ELISA. All steps were performed at room temperature (18-22°C). Samples of purified antibodies containing variant Fc regions were diluted to 10 μg / mL in PBS. The negative control consisted of PBS alone. 100 μL of each replicate was added to two 96-well clear Maxisorp microplates (Sigma catalog number M9410-ICS) and incubated with shaking for approximately 60 minutes. The plates were rinsed with PBS containing 0.05% (v / v) Tween 20 (wash buffer), and 200 μL of blocking buffer was added. The plates were incubated with shaking for approximately 60 minutes, optionally stored overnight at 4°C, and then rinsed with wash buffer. HRP-labeled C1q was diluted to 100 ng / mL in block buffer, and 100 μL was added to each well. The plates were incubated with shaking for approximately 90 minutes, then washed four times with wash buffer and twice with water. 100 μL of 3,3',5,5'-tetramethylbenzidine liquid substrate, ultrasensitive (TMB) (Sigma catalog number T4444) was added, and the plate was incubated with shaking for approximately 10 minutes. The reaction was stopped by adding 50 μL of 1 M sulfuric acid, and absorbance was read at 450 nm minus 620 nm using a microplate spectrophotometer (Anthos Labtec HT). The entire experiment was repeated three times using different selections of variant antibodies. In experiments 1 and 2, six replicates of each sample were tested (three on each plate), and in experiment 3, 12 replicates were tested (six on each plate). The mean absorbance for each sample was calculated, as shown in Table 17. Within each experiment, the mean absorbance for each sample was compared to the mean absorbance obtained with buffer alone using a one-tailed t-test assuming unequal variances. The resulting mean probability (P) values ​​were calculated. Results were marked if they were significantly higher than the response of buffer alone (p≦0.05). The results are shown in Table 17. With the exception of 2-25 (L234K / L235T / G236R) and 2-39 (L234R / L235I / G236R), all of the new variants (samples 2-2 through 2-63) and controls (2-65, 2-66, and 2-67) were indistinguishable from buffer alone in binding to C1q.

[0247] [Table 18-1]

[0248] [Table 18-2]

[0249] The above experiments were performed using variants with variable regions derived from the anti-CD20 antibody rituximab. To demonstrate the generality of the above results, additional experiments were performed using variants with variable regions derived from the anti-CD3 antibody muromonab (OKT3) or the anti-CD52 antibody alemtuzumab, as well as other variants based on different immunoglobulin species and isotypes.

[0250] Binding analysis was performed by ELISA following essentially the same method as previously described. Alternative sources of microplates (Sigma catalog no. CLS 9018) and TMB substrate (Abcam catalog no. ab171523), as well as variations such as extended incubation (overnight) with HRP-labeled C1q, did not substantially alter the results. Samples were tested in triplicate in two microplates (six replicates total). The mean absorbance of each sample was compared to the mean absorbance obtained with buffer alone using a one-tailed t-test assuming unequal variances, and the resulting probability (P) values ​​were calculated. Results were marked if they were significantly higher than the response of buffer alone (p≦0.05). Experiments were performed in duplicate. Table 18 shows the results, along with the mean P values. The results for the variant anti-CD3 and anti-CD52 antibodies were consistent with previous results for the variant anti-CD20 antibody. Their binding to C1q was not significantly greater than buffer alone (p>0.05), except for variants 4-63 (anti-CD52 L234T / L235T / G236R), 4-65 (anti-CD52 L234A / L235A), and 4-66 (anti-CD52 L234A / L235A / P329G), which yielded responses very slightly greater than those of buffer alone. Furthermore, there was no significant difference between the binding of the new variants to C1q and that of the corresponding LALAPG reference. For human IgG2 with the alteration L234Q / L235S / Δ236R and human IgG4 and rabbit IgG with the alteration L234Q / L235S / G236R, binding to C1q was not significantly greater than buffer alone. Human IgG2 with only the two changes L234Q / L235S showed slightly increased binding (p=0.017), as did rat IgG2b with the three changes L234Q / L235S / G236R (p=0.008). In contrast, mouse IgG2a with the same three changes L234Q / L235S / G236R resulted in levels of C1q binding equivalent to wild-type human IgG1. This is consistent with previous observations showing that the binding site for C1q on mouse IgG is different from that on human IgG (Idusogie 2000).Finally, the combination of the alterations L234Q / L235S / G236R with two different sets of alterations that result in increased binding to FcRn, namely M252Y / S254T / T256E or T250Q / M428L, still resulted in binding to C1q that was not significantly greater than buffer alone.

[0251] [Table 19-1]

[0252] [Table 19-2]

[0253] The results from this set of experiments demonstrate the generality of the amino acid changes we discovered to reduce C1q binding to undetectable levels in a wide variety of settings, e.g., different variable regions, different IgG subclasses, different species contexts, etc., and also in combination with other amino acid changes commonly introduced into IgG.

[0254] Example 5 In silico evaluation of potential binding to MHC class II The potential of variant Fc regions to yield peptides with theoretical binding capacity to human major histocompatibility complex class 2 (MHC class II) was assessed using the MHC-II binding prediction tool provided by the Immune Epitope Database (IEDB) (Fleri 2017), available at http: / / tools.iedb.org / mhcii / (accessed October 17, 2019). The prediction method was version 2.22, recommended by the IEDB. This uses the consensus method, version 2.22, combining NN-Align (Jensen 2018), SMM-Align (Nielson 2007), CombLib (Sidney 2008), and Sturniolo (Sturniolo 1999), if any corresponding predictor variables were available for the molecule; otherwise, NetMHCIIpan (Jensen 2018) was used.

[0255] The input consisted of a list of 361 peptides in FASTA format corresponding to amino acid residues C220 to T250 (inclusive) of human IgG1, with every residue at positions 234 and 235 other than C, and position 236 substituted with R. A "7-allele" reference set of MHC class II alleles was selected (Paul 2015). This consisted of the following alleles: DRB1*03:01, DRB1*07:01, DRB1*15:01, DRB3*01:01, DRB3*02:02, DRB4*01:01, and DRB5*01:01. A default peptide epitope length of 15 amino acid residues was selected. The output was sorted by rank score and presented as an XHTML table. Data were exported to a Microsoft Excel file for further processing. AnFc variants were considered "at risk" of generating novel peptides with potential MHC class II binding if any one of the potential 15-mer peptides that could result from the 31-residue input peptide yielded a rank score of 10% or less for binding to any of the selected alleles. The entire process was repeated using the full reference set of 27 MHC class II alleles (Greenbaum 2011). The results are shown in Table 19. The presence of hydrophobic amino acid residues F, I, L, M, V, W, or Y at position 234 and / or F, W, or Y at position 235 was associated with an increased risk of generating novel peptides with potential MHC class II binding and therefore potentially unwanted immunogenicity.

[0256] [Table 20]

[0257] In this table, a "7" indicates that at least one peptide produced a rank score <10% with one or more alleles from the 7-allele reference set, an "A" indicates that at least one peptide produced a rank score <10% with one or more alleles from the 27-allele reference set, and a blank indicates that no peptides produced a rank score <10% with any allele.

[0258] Example 6 Activity in ADCC and ADCP 6.1 ADCC using FcγRIIIA Antibodies were evaluated for their ability to engage in ADCC using the ADCC Reporter Bioassay (Promega Catalog No. G7015). This kit contains Raji target cells expressing the CD20 antigen and engineered Jurkat effector cells stably expressing the FcγRIIIA receptor, the V158 (high affinity) variant, and the nuclear factor of activated T cells (NFAT) response element, which can drive the expression of firefly luciferase. This allows for sensitive measurement of an early response in the ADCC pathway, namely, activation of NFAT-mediated gene transcription. Experiments were performed according to the manufacturer's instructions. Target cells (Promega Catalog No. G960A), effector cells (Promega Catalog No. G701A), and sample dilutions were all prepared in RPMI 1640 culture medium (Promega Catalog No. G708A) containing 4% low IgG serum (Promega Catalog No. G711A) (assay buffer). 75 μL of assay buffer was added to the edge wells of a white, flat-bottom microplate (Costar Catalog No. 3912), and 25 μL of target cell suspension was added to the center well. Sample dilutions were prepared offline in the plate as needed; 25 μL was transferred to the center well of the assay plate and mixed on a plate shaker. 25 μL of effector cell suspension was added, mixed, and the plate was incubated at 37°C and 5% CO2 for approximately 6 hours, then equilibrated to room temperature for approximately 15 minutes. Luciferase assay substrate (Promega Catalog No. G720A) was reconstituted by adding luciferase assay buffer (Promega Catalog No. G719A) equilibrated to room temperature, and 75 μL was added to the center well of the microplate. Luminescence was measured within 10 minutes using a microplate luminometer (Promega Glomax® 96).

[0259] The entire experiment was performed twice, using two microplates per experiment. On each plate, the wild-type reference anti-CD20 antibody (Sample 2-1) was titrated starting at a concentration of 3 μg / mL (1 μg / mL final in the assay) and subsequently diluted 2.5-fold to determine the dose response. The dose-response curves are shown in Figure 3 , which demonstrate sufficient assay performance. Test samples were diluted to 30 μg / mL (10 μg / mL final in the assay). All control and test samples were tested in duplicate. Normalized responses were expressed as a percentage of the response of the wild-type reference antibody at 1 μg / mL final in the same microplate. The mean normalized response for each sample across four replicates was compared to the mean normalized response of assay buffer using a one-tailed t-test assuming unequal variances. Results were flagged if significantly higher than the normalized response of assay buffer (p≦0.05). The results are shown in Table 20. With the exception of 2-25 (L234K / L235T / G236R) and 2-36 (L234R / L235D / G236R), all of the new variants (including samples 2-2 through 2-64, respectively) produced responses indistinguishable from assay buffer. In contrast, the previously published variants 2-65 (L234A / L235A), 2-75 (L234F / L235E / P331S), and 2-76 (L234F / L235Q / K322Q) produced significantly higher responses compared to assay buffer.

[0260] [Table 21-1]

[0261] [Table 21-2]

[0262] 6.2 ADCP using FcγRIIA Antibodies were evaluated for their ability to engage ADCP using an ADCP reporter bioassay (Promega catalog no. G9901). This kit contains Raji target cells expressing the CD20 antigen and engineered Jurkat effector cells stably expressing the FcγRIIA receptor, H131 variant, and the nuclear factor of activated T cells (NFAT) response element, which can drive the expression of firefly luciferase. The assay was performed using FcγRIIA-H effector cells (Promega catalog no. G988A) according to the manufacturer's instructions and as described in Example 6.1. Experiments were performed once, with each sample tested in duplicate. Normalized responses were calculated as described in Example 6.1. Figure 4 shows the dose-response curve for titration of the wild-type reference antibody (Sample 2-1). Table 21 shows the results. With the exception of 2-25 (L234K / L235T / G236R), 2-30 (L234Q / G236R), 2-36 (L234R / L235D / G236R), and 2-58 (L234T / L235K / G236R), all of the new variants (including samples 2-2 through 2-64, respectively) yielded responses that were not significantly different from assay buffer. In contrast, the previously published variants 2-65 (L234A / L235A), 2-75 (L234F / L235E / P331S), and 2-76 (L234F / L235Q / K322Q) yielded significantly higher responses compared to assay buffer.

[0263] [Table 22-1]

[0264] [Table 22-2]

[0265] Example 7 Size exclusion chromatography The purified antibody was analyzed by size exclusion chromatography (SEC). SEC was performed on an Agilent 1100 series HPLC using a Superdex 200 Increase 5 / 150GL column equilibrated in PBS, pH 7.2, at a flow rate of 0.2 mL / min. Absorbance was measured at 280 nm, and the area of ​​the peak corresponding to IgG monomer was measured as a percentage of the total protein eluted. Samples were analyzed immediately after preparation and again at 7 and 14 days after incubation at 40°C. The results are shown in Table 22. On average, there was a decrease of approximately 1.7% in the monomer fraction after 7 days at 40°C and approximately 4.4% after 14 days, which was closely mirrored by the wild-type reference sample.

[0266] [Table 23-1]

[0267] [Table 23-2]

[0268] [Table 23-3]

[0269] Example 8 Differential scanning fluorimetry Antibodies were evaluated using differential scanning fluorimetry (DSF) to provide an indication of their relative thermal stability using the Protein Thermal Shift Dye Kit (ThermoFisher catalog number 4461146) according to the manufacturer's instructions. Five microliters of Protein Thermal Shift Dye buffer was added to a polypropylene microplate (ThermoFisher catalog number N8010560). Ten microliters of test sample (1 mg / mL in PBS) was added and mixed. Protein Thermal Shift Dye (1000x) was diluted with water to obtain an 8x intermediate stock, and 2.5 µL was added to each sample and mixed. Melting curves were measured using a real-time PCR system (ThemoFisher QuantStudio 3) using an Orange filter set with excitation at 580 ± 10 nm and emission at 623 ± 14 nm. The temperature was increased in successive steps from 25°C to 90°C at a rate of 0.05°C / s. Measurements were taken at intervals of approximately 0.1°C. Because IgG molecules contain several different domains that are expected to melt at different temperatures, the fluorescence response did not, as expected, exhibit a single, clearly defined melting temperature (Tm). To obtain a measure of the relative thermal stability of the different variants, the temperature at which the fluorescence was consistently 2000 units higher (over three consecutive measurements) than the average baseline fluorescence measured between 25°C and 35°C was determined. This temperature was interpreted as the onset of the melting (denaturation) process. Each sample was measured in duplicate. The average results are shown in Table 23. With the exception of sample 2-67 (aglycosyl), all of the variants began to melt at temperatures within the range of 61.2°C to 67.6°C, which was 2.6°C lower and 3.8°C higher than the wild-type reference sample 2-1.

[0270] [Table 24-1]

[0271] [Table 24-2]

[0272] Example 9 Cytokine release Antibodies were evaluated for their ability to induce cytokine release from human peripheral blood mononuclear cells (PBMCs). PBMCs were isolated from five healthy donors by centrifugation over Ficoll-Paque PLUS (GE Healthcare catalog number 11778538) and washed with RPMI medium (Sigma catalog number 8758) containing 1% heat-inactivated fetal bovine serum (FBS) (Fisher Scientific catalog number 11550526) and 1% penicillin / streptomycin (PS) (Sigma catalog number 4333). PBMCs were cultured at 2 x 10 in RPMI containing 10% FBS and 1% PS (cRPMI). 6 Anti-CD3 antibodies were resuspended at 200, 20, 2, 0.2, or 0.02 μg / mL. Anti-CD3 antibodies at 1 mg / mL in PBS were diluted with cRPMI to obtain concentrations of 200, 20, 2, 0.2, or 0.02 μg / mL. The negative control consisted of PBS diluted 5-fold with cRPMI. Each sample was tested in triplicate. 100 μL of the test sample was mixed with 100 μL of the cell suspension in a 96-well round-bottom microplate and cultured at 37°C and 5% CO2 for 24 hours. The plate was centrifuged at 400 × g for 5 minutes, and the culture supernatant was collected. Cytokines (GM-CSF, IFNγ, IL-2, IL-4, IL-6, IL-8, IL-10, and TNFα) were quantified in the cell culture supernatant by Luminex assay (Bio-Rad catalog number M50000007A) on a Bio-Plex® 200 system with Bio-Plex Manager software according to the manufacturer's instructions. For each cytokine, the mean concentrations obtained from all five donors were compared to the mean concentrations obtained with the negative control using a one-tailed t-test assuming unequal variances. Results were marked if they were significantly higher than the response with the negative control (p ≤ 0.05). Positive responses were obtained with the wild-type reference antibody at final concentrations of 10, 1, 0.1, and 0.01 μg / mL.

[0273] Figure 5 shows cytokine release results from 10 μg / mL in five donors stimulated at a final concentration of 10 μg / mL. The variants L234S / L235T / G236R, L234S / L235V / G236R, and L234A / L235A / P329G (LALAPG) did not result in a significant increase over the negative control for any of the cytokines. Similar results were obtained from five independent donors stimulated at 0.1 μg / mL.

[0274] Example 10 Pharmacokinetics Tg276 transgenic mice express the human FcRn gene under the control of a strong CAG promoter (Proetzel 2014). The mice were backcrossed to a congenic FcRn-null strain so that the strain expresses high levels of human FcRn but does not express mouse FcRn. These mice are suitable for detecting subtle differences in antibody persistence in vivo.

[0275] Pharmacokinetic (PK) studies were performed by Jackson Laboratory (Bar Harbor, ME, USA) according to their standard procedures. Twenty-four 10-week-old male B6.Cg-Fcgrttm1Dcr Tg(CAG-FCGRT)276Dcr / DcrJ (Tg276, JAX stock number 004919) homozygous mice were distributed into four groups with six mice per group. On day 0, all mice received an IV dose of the test article at 5 mg / kg. Blood samples were collected in EDTA at 15 minutes, 3 hours, 1 day, 3 days, 5 days, 7 days, 10 days, and 14 days, processed to plasma, and stored frozen until analysis. Plasma samples were diluted 1 / 500 and evaluated by human IgG Fc-specific ELISA using each test article as a standard. Dose-response curves were fitted by a five-parameter logistic equation, and test sample concentrations were determined by interpolation. Each test sample was measured in triplicate, and the mean concentration was used to calculate kinetic parameters. Data were fitted to a two-phase model by noncompartmental analysis, and elimination half-lives were calculated from 24 hours to 15 days. The results are shown in Table 24. Differences in half-lives between samples were analyzed by Tukey's multiple comparison test. There were statistically significant differences between each of the variants and the wild-type, but no statistically significant differences between the variants. All pharmacokinetic parameters for each of the variants (half-life, clearance, Cmax, area under the curve, and volume of distribution) were within ±50% of the corresponding parameters for the wild-type antibody.

[0276] [Table 25]

[0277] Example 11 Glycosylation N-linked carbohydrates were analyzed after separation by hydrophilic interaction liquid chromatography (HILIC). N-glycans were released using Rapid PNGase F (NEB catalog no. P07010) essentially according to the manufacturer's instructions. Released glycans were fluorescently labeled by reductive amination with 2-aminobenzoic acid and sodium cyanoborohydride. Labeled oligosaccharides were purified by HILIC using DPA-6S solid-phase extraction cartridges (Sigma catalog no. 52624-U) as described (Neville 2004). Purified labeled oligosaccharides were diluted with acetonitrile (final concentration 66% acetonitrile) and analyzed using an Agilent 1260 Quaternary HPLC system with fluorescence detection (excitation at 240 nm, emission at 430 nm) using a BEH amide column, 3.5 μm, 2.1 mm x 150 mm (Waters catalog no. 186004861) at 30 °C. The column was eluted at 0.15 mL / min with a gradient of 85:10:5 to 50:15:35 acetonitrile:0.525 M ammonium acetate pH 3.75:water. Figure 6A shows a representative chromatogram of each sample. The same glycoform spectrum was observed for each sample. The experiment was repeated using selected CD20 antibodies prepared from both HEK and CHO cells. Figure 6B shows a representative chromatogram. The glycoform spectra differed slightly between antibodies prepared from different cell types, but within each cell type, the spectra were identical.

[0278] Example 12 thermal stability Antibodies were evaluated using the Uncle® Biostability Platform (Unchained Labs, Pleasanton, CA, USA) according to the manufacturer's instructions. Samples were subjected to a thermal gradient experiment, starting at 25°C and increasing to 95°C at a rate of 0.3°C / min. Intrinsic protein fluorescence and static light scattering (266 nm) were measured throughout the experiment as indicators of protein unfolding and aggregation, respectively. Each sample was tested at a concentration of 1 mg / mL in phosphate-buffered saline (PBS) and in a sample volume of 9 μL. 48 individual samples were tested simultaneously, and the entire experiment was performed in triplicate to obtain triplicate independent temperature profiles for each test sample. The "melting temperature" Tm was defined as the major inflection point in the fluorescence vs. temperature plot, i.e., the temperature at which the rate of change in fluorescence reached a maximum. The "aggregation temperature" Tgg was defined as the temperature at which light scattering began to increase. Variants were compared to the wild-type reference by two-tailed t-test. The results are shown in Table 25.

[0279] The aglycosyl variants of the CD20 and CD3 antibodies had significantly lower Tm, and all other variants had Tm that was higher or not significantly different from wild-type. The aglycosyl variants of the CD20 and CD3 antibodies and the L234Q / L235S / G236R CD20 variant had significantly lower Tagg, and all other variants had Tagg that was higher or not significantly different from wild-type.

[0280] [Table 26]

[0281] Example 13 Assessment of unwanted immunogenicity in vitro The ProMap® Immunogenicity System is based on a flow cytometry method that analyzes cell proliferation by tracking cells labeled with carboxyfluorescein succinimidyl ester (CFSE). Protein or peptide antigens presented by antigen-presenting cells to CD4+ T cells result in cell proliferation that can be measured by a decrease in the fluorescence intensity of CFSE-labeled cells. The study was performed by Proimmune (Oxford, UK) according to their standard Promap® T cell proliferation assay procedure. Test samples consisted of purified 20-mer peptides (Kendall Scientific Inc., Lincolnshire, IL, USA). The sequences of the peptides are shown in Table 26.

[0282] [Table 27]

[0283] The positive control samples performed as expected. Surprisingly, peptide 09 (G236R) produced positive responses in 4 of 20 donors. Peptide 11 (L234S / L235V / G236R) produced positive responses in 2 of 20 donors, while the other peptides produced positive responses in none or only one donor. If a peptide is considered a potential T cell epitope, at least two donors will produce a positive response.

[0284] [References] Angal 1993 Mol. Immunol. 30:105-108. Armor 1999 Eur. J. Immunol. 29:2613-2624. Bajaj 2012 J. Appl. Pharm. Sci. 2:129-138 Beck 2011 mAbs 3:415-416. Boesch 2014 mAbs 6:915-927. Bolt 1993 Eur. J. Immunol. 23:403-411. Borrok 2017 J. Pharm. Sci. 106:1008-1017. Brinkmann 2017 MAbs 9:182-212. Brinks 2013 Pharm. Res. 30:1719-1728. Bruhns 2009 Blood 113:3716-3725. Bryson 2010 Biodrugs 24:1-8. Chen 2007 Retrovirology 4:33. Choi 2013 Mol. Cancer Ther. 12:2748-2759. Choi 2015a Mol. Immunol. 65:377-383. Choi 2115b PLoS One 10:e0145349. Costa 2014 Crit. Rev. Biotechnol. 34:281-299. Dall’Acqua 2006 J. Immunol. 177:1129-1138. Datta-Mannan 2011 Drug. Metab.. Dispos 42:1867-1872. de Jong 2016 PLoS Biol 14:e1002344. Deng 2010 Drug Metab. Dispos. 38:600-605. Dhanda 2019 Nucleic Acids Research 47:W502-W506. Edelman 1969 Proc. Natl. Acad. Sci. USA 63:78-85. Engelsman 2011 Pharm. Res. 28:920-933. Fan 2014 Mol. Cancer Therapeutics 2014:0715. Finco 2014 Cytokine 66:143-155. Findlay 2010 J. Immunol. Meth. 352:1-12. Findlay 2011 Cytokine 55:141-151. Fleri 2017 Frontiers in Immunology 8:278. Greenbaum 2011 Immunogenetics. 63:325-35, 2011. Gregorio 2019 Methods Protocols 2:24. Grimaldi 2016 Cytokine 85:101-108. Gunasekaran 2010 J. Biol. Chem. 285:19637-19646. Hale 1983 Blood 62:873-882. Hezarah 2001 J. Virol. 75:12161-12168 Hinton 2004 J. Bio. Chem. 279:6213-6216. Horton 2011 J. Immunol. 186:4223-4233. Hsiao 2018 Breast Cancer Research 20:43. Kabat 1991 Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD. Kim 1999 Eur. J. Immunol. 29:2819-2825. Idusogie 2000 J. Immunol. 164:4178-4184. Ishino 2013 J. Biol. Chem. 288:16529-16537. Jafari 2017 Curr. Med. Chem. 24:1228-1237. Jawa 2013 Clinical Immunology 149:534-555. Jefferis 2005 Biotechnol. Prog. 21:11-16. Jensen 2018 Immunology 154:394-406. Joubert 2016 PLoS ONE 11(8):e0159328. Kanda 2006 Glycobiology 17:104-118. King 2014 Proc. Natl. Acad Sci. 111:8577-8582. Kuo 2011 mAbs 3:422-430. Labrijn 2013 Proc. Natl. Acad. Sci.110:5145-5150. Lazar 2006 Proc. Natl. Acad. Sci.103:4005-4010 Leaver-Fay 2016 Structure 24:641-651. Lee 2019 Nature Communications 10:5031. Lin 2013 Cancer Biology & Therapy 14:1032-1038 Liu 2018 Protein Cell 9:15-32. Loureiro 2011 J. Virology 85:3010-3014. Lund 1996 J Immunol. 157:4963-4969. Mekhaiel 2011 Sci. Rep. 1:124. Merchant 1998 Nat. Biotechnol. 16:677-681. Michaelsen 2009 Scand. J. Immunol. 70:553-564. Moore 2010 mAbs 2:181-189. Moore 2011 mAbs 3:546-557. Moore 2019 Methods 154:38-50. Nardis 2017 J. Biol. Chem. 292:14706-14717. Neville 2004 Anal. Biochem. 331:275-282. Nielson 2007 BMC Bioinformatics 8:238. Niesen 2007 Nature Protocols 2:2212-2221. Niles 2007 Anal. Biochem. 366:197-206. Oganesyan 2008 Acta Cryst. D64:700-704. Paul 2015 J. Immunol. Meth. 422:28-34. Petkova 2006 Int. Immunol. 18:1759-1769. Proetzel 2014 Methods 65:148-153. Remington 1995 The Science and Practice of Pharmacy, 19th Edition, Gennaro, Ed., Mack Publishing Co. Reutsch 2015a mAbs, 7:167-179. Reutsch 2015b mAbs, 7:732-742. Ridgway 1996 Protein Eng. Design 9:617-621. Romer 2011 Blood 118:6772-6782. Rose 2011 Structure 19:1274-1282. Rowland 1995 Clinical Pharmacokinetics: Concepts and Applications, Third Edition, pub Lippincott, Williams & Wilkins. Schlothauer 2016 Protein Engineering, Design & Selection 29:457-466. Senisterra 2009 Mol. BioSyst. 5:217-223. Shan 2016 PLoS One 11:e0180345. Shields 2011 J . Biol. Chem. 276:6591-6604 Shire 2009 Current Opin. Biotechnol. 20:708-714. Sidney 2008 Immunonome Research 4:2. Simeon 2018 Protein Cell 9:3-14. Stapleton 2018 Mol. Immunol. 95:1-9. Strop 2012 J. Mol. Biol. 420:204-219. Sturniolo 1999 Nat. Biotechnology 17:555-561. Tam 2017 Antibodies 6:1-34. Vaccaro 2005 Nat. Biotechnol. 23:1283-1288. Vafa 2014 Methods 65:114-126. Vessilier 2015 J. Immunol. Meth. 424:43-52. Von Kreudenstein 2013 mAbs 5:646-654. Wang 2008 PLoS Comput Biol. 4:e1000048. Werner 2007 Acta Paediatrica 96:17-22. Wilkinson 2013 mAbs 5:406-417. Wing 1995 Therapeutic Immunology 2:183-190. Wolf 2012 Cytokine 60:828-837. Wozniak-Knopp 2017 Protein Eng. Design 30:657-671. Xu 2000 Cell. Immunol. 200:16-26. Yang 2018 Frontiers in Immunology 8:1860. Yeung 2009 J Immunol 2009; 182:7663-7671. Ying 2012 J. Biol. Chem. 287:19399-19408. Ying 2014 mAbs 6:1201-1210. Zalevsky 2010 Nat. Biotechnol 28:157-159. Zhou 2017 Biomedicines 5:64.

Claims

1. 1. A fusion protein comprising a variant IgG Fc region or variant Cγ2 domain, wherein the variant IgG Fc region or variant Cγ2 domain is either (a) a variant Fc region or variant Cγ2 domain comprising the amino acid substitutions L234S / L235T / G236R; (b) a variant human IgG2 Fc region or variant IgG2 Cγ2 domain comprising the amino acid changes V234S / A235T / Δ236R; or (c) a variant human IgG4 Fc region or variant IgG4 Cγ2 domain comprising the amino acid substitutions F234S / L235T / G236R, wherein the amino acid numbering is according to the EU index as set forth in Kabat.

2. A conjugate comprising the fusion protein of claim 1.

3. A composition comprising the fusion protein of claim 1 or the conjugate of claim 2.

4. A medicament for treating a mammal, comprising the fusion protein of claim 1; or the conjugate of claim 2; or the composition of claim 3.

Images