HK1202565B - Sequence symmetric modified igg4 bispecific antibodies - Google Patents

Description

IgG4 bispecific antibodies with modified sequence symmetry

The present disclosure relates to asymmetric antibodies comprising a mutated IgG4 heavy chain or fragment and a second heavy chain or fragment different from the IgG4 chain. The present disclosure also extends to compositions comprising the asymmetric antibodies and the use of the antibodies and compositions comprising the antibodies for therapeutic purposes. In other aspects, the present disclosure extends to methods of preparing the antibodies and formulations, as well as vectors encoding the antibodies and hosts for expressing the same.

The biopharmaceutical industry, which includes recombinant proteins, monoclonal antibodies (mAbs), and nucleic acid-based drugs, is growing rapidly. Antibody engineering has led to the design and generation of antibody fragments or alternative forms. When selecting an antibody-based protein as a therapeutic agent, the preferred molecular form is considered together with other aspects such as product yield, protein quality, and storage stability.

The basic structure of all immunoglobulin (Ig) molecules consists of two identical heavy chains (HC) and two identical light chains (LC) coupled by disulfide bonds. Each LC is composed of a variable domain ( _VL ) and a constant domain ( _CL ). Based on the HC, five major Ig classes are recognized: IgG, IgA, IgD, IgE, and IgM. For IgG, the HC is composed of one variable domain ( _VH ) and three constant domains _{( CH1-3} ). _{The CH2} and _CH3 domains form the Fc portion of the molecule, which is responsible for stimulating effector functions and is connected to the Fab fragments ( _VHVL and _CHCL ) _by the hinge region that imparts flexibility to the IgG molecule. Two antigen recognition sites are located at the ends of the _VL and _VH domains. IgG is further subdivided into four different isotypes: IgG1, IgG2, IgG3, and IgG4.

The effector functions mediated by Fc-, i.e., antibody-dependent cellular cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC), are isotype-dependent. Each isotype has evolved to perform a specific function in vivo. The IgG1 isotype is currently the most widely used therapeutic agent due to its extended half-life, enhanced ADCC activation, and complement activation. Depending on the target and desired effect, other isotypes are used as therapeutic agents. For example, when only neutralizing the target antigen and the effector function is less important, alternative isotypes such as IgG2 and IgG4 can be used. Alternatively, IgG with reengineered Fc/effector functions can be considered.

IgG2 also has minimal associated effector function but is prone to dimerization, a phenomenon that is not fully understood.

IgG4 remains a useful isotype due to its relative lack of effector function induction. However, the use of IgG4 also has some inherent practical difficulties, namely its shorter serum half-life and its ability to undergo "Fab-arm exchange", also known as dynamic heavy chain exchange or heavy chain exchange, in which the heavy chain of one antibody and its associated light chain are exchanged with the heavy chain of another antibody and its associated light chain, thereby forming a complete antibody consisting of two heavy chains and two associated light chains (van der Neut Kolfschoten et al., 2007 Science 317, 1554-1557).

In vivo, Fab-arm exchange results in bispecific antibodies that, due to their different variable domains, can simultaneously co-engage different target antigens. This results in a high percentage of circulating IgG4, which has been observed to be bispecific but functionally monovalent (Schuurman, J., Van Ree, R., Perdok, G.J., Van Doorn, H.R., Tan, K.Y., Aalberse, R.C., 1999. Normal human immunoglobulin G4 can be bispecific: it has two different antigen-combining sites. Immunology 97, 693-698).

In vitro, when IgG4 antibodies are analyzed by non-reducing SDS-PAGE, they have been observed to form so-called "half-molecules," each comprising a single covalently associated heavy chain-light chain pair due to the absence of inter-heavy chain disulfide bonds (typically resulting from the formation of intra-heavy chain disulfide bonds within the hinge region of one heavy chain). The heavy chain of a "half-molecule" can non-covalently associate with its heavy chain partner, with the association being maintained by CH3:CH3 domain interactions. In solution, such "half-molecules" have been observed to be fully sized, i.e., approximately 150 kDa, using methods such as size exclusion chromatography, but on non-reducing SDS-PAGE, they comprise a 75 kDa LC:HC pair (the so-called "half-molecule").

The mutation of Ser to Pro at position 241 (numbered according to the Kabat numbering system) in the hinge reduces the appearance of these "half molecules" as shown by non-reducing SDS-PAGE (Angal, S. et al., 1993. A single amino acid substitution abolishes the heterogeneity of chimeric mouse/human (IgG4) antibody as observed during SDS-PAGE analysis Mol Immunol 30, 105-108). In addition, this point mutation does not affect the compact structure of IgG4, allowing IgG4 to retain its reduced ability to activate complement.

Following the discovery of the S241P mutation, other IgG4 mutations have been investigated to understand inter-heavy chain interactions of IgG4 antibodies, reduce IgG4 effector function, and enhance structural stability. In Schuurman et al. (Schuurman, J et al., 2001. The inter-heavy chain disulphide bonds of IgG4 are in equilibrium within intra-heavy chain disulphide bonds. Molecular Immunology 38, 1-8), IgG4 mutants were used to investigate the observed instability of the inter-heavy chain disulfide bonds of IgG4. In mutant M1, Cys131 (numbered according to the EU numbering system or Cys127 according to the Kabat numbering system), involved in the heavy chain-light chain ( _CL - _CH1 ) disulfide bond, was replaced with serine, and this mutant was found to result in the formation of both light chain and heavy chain dimers. In mutant M2, cysteine 226 (226 according to the EU numbering system or 239 according to the Kabat numbering system), which participates in the inter-heavy chain disulfide bond in the hinge, was replaced by serine, and this mutant was found to have a more stable inter-heavy chain connection than IgG4 and prevent the formation of intra-heavy chain disulfide bonds.

Previously, changes in the number of cysteine residues present in the hinge region of antibodies have been studied. US5677425 Bodmer et al. disclose that the number of cysteine residues in the hinge region can be increased to facilitate the use of cysteine sulfhydryl groups for linking effectors or reporter molecules. US5677425 also teaches that the number of cysteine residues in the hinge region can be reduced to one to facilitate the assembly of antibody molecules, because only a single disulfide bond will need to be formed, which will provide a specific target for linking the hinge region to another hinge region or to an effector or reporter molecule.

Given that IgG4 antibodies administered to a subject are susceptible to dynamic heavy chain exchange to form "hybrid antibodies," this process can be utilized in the present disclosure to prepare antibodies of the present disclosure in vitro. Advantageously, this allows the characteristics of the antibody to be manipulated.

There remains a need to provide novel antibodies for use as therapeutic agents. The present invention provides novel mutant antibodies that may have advantageous properties, including, for example, improved biophysical properties compared to wild-type antibodies.

SUMMARY OF THE INVENTION

The present disclosure provides asymmetric hybrid antibodies comprising two heavy chains or heavy chain fragments each comprising at least a variable region, a hinge region and a _CH1 domain,

wherein the first heavy chain or fragment thereof is of class IgG4 and has:

a) the interchain cysteine at position 127 according to the Kabat numbering system in the _CH1 domain is replaced by another amino acid; and

b) Optionally, one or more amino acids located in the upper hinge region are replaced by cysteine.

The second heavy chain or fragment thereof is characterized in that part or all of the entire chain has an amino acid sequence different from that of the first heavy chain in at least the region outside the variable region.

In an alternative aspect, the disclosure relates to an asymmetric hybrid antibody comprising an IgG4 heavy chain or heavy chain fragment, wherein the heavy chain or fragment comprises a variable region, a hinge region and a _CH1 domain, wherein the hinge is mutated to an IgG1-type hinge.

Also provided are asymmetric hybrid antibodies comprising a first and a second heavy chain or heavy chain fragment each comprising at least a variable region, a hinge region and a _CH1 domain, wherein the first heavy chain or fragment thereof is of class IgG4 and has an IgG1-type hinge, and the second heavy chain or fragment thereof has an amino acid sequence that differs from that of the first heavy chain in a region outside of at least the variable region.

In one embodiment, the hinge regions of the two heavy chains are similar or identical.

The present disclosure is advantageous because it allows manipulation and control of antibody properties through routine and readily available methods.

The antibodies of the present invention can exhibit reduced heavy chain exchange compared to wild-type IgG4, which provides asymmetric antibodies (e.g., bispecific antibodies) that show little or no exchange with wild-type IgG4 in vivo (due to their reduced propensity for exchange compared to IgG4 and also due to relatively low in vivo concentrations of asymmetric mixed antibodies compared to naturally circulating IgG4 antibodies).

The antibodies of the present invention can exhibit reduced heavy chain exchange compared to IgG4 wild type at concentrations higher than in vivo concentrations, for example, 0.5 mM or higher. Although the antibodies of the present invention exhibit reduced heavy chain exchange compared to wild type IgG4, they do exhibit a certain degree of heavy chain exchange compared to IgG1wt and IgG4S241P, which is sufficient to generate asymmetric hybrid antibodies in vitro from two different antibodies (e.g., two antibodies with different antigenic specificities).

Thus, in one embodiment, the antibodies of the present disclosure can be exchanged in vitro using 5 mM GSH rather than 0.5 mM GSH. The latter more closely mimics the in vivo exchange barrier. Figure 20 illustrates this fact, showing that Ab28 (C127S Y229C) does exchange with wild type, S241G S241A, or S241T at 5 mM GSH, but not at 0.5 mM.

Therefore, the present invention also provides a method for producing an asymmetric hybrid antibody, comprising the following steps: obtaining a symmetric antibody comprising a first heavy chain sequence or a fragment thereof as defined herein, mixing said antibody with a second symmetric antibody comprising a second heavy chain sequence or a fragment thereof different from the first heavy chain sequence in vitro under conditions that favor heavy chain exchange between the two antibodies, and optionally isolating the asymmetric hybrid antibody obtained therefrom. In one embodiment, the method provides a bispecific antibody comprising mixing two antibodies, wherein the antigenic specificity of the variable region of the first antibody is different from the antigenic specificity of the variable region of the second antibody.

In one embodiment, the antibody is monovalent.

The methods of the present disclosure allow for complete manipulation of the properties of antibodies to provide ultimate therapeutic molecules that are tailored and optimized for the desired therapeutic use.

Furthermore, antibodies according to the present disclosure may be advantageous in that they have low levels of effector function and/or do not participate in cross-linking.

Summary of the Figures

Figure 1 a shows the human _CH1 and hinge sequences of IgG1 wild type and IgG4 wild type, with hinge residues underlined, and the kappa light chain constant sequence.

Figure 1b shows:

Human kappa light chain constant sequence with the cysteines forming the interchain _CL - _CH1 disulfide bond indicated (underlined);

Sequences of the N-terminal _CH1 residues and hinge regions of human IgG1, 2, 3, and 4 heavy chains, with the positions of the cysteines forming interchain _CL - _CH1 disulfide bonds indicated (underlined) (in the upper hinge for IgG1 and in the N-terminal _CH1 for IgG2, 3, and 4);

The N-terminal _CH1 residues and portion of the hinge region sequence of the human IgD heavy chain, with the positions of the cysteines in the N-terminal _CH1 sequence that form the interchain _CL - _CH1 disulfide bond indicated (underlined);

the N-terminal _CH1 , C-terminal _CH1 residues, and selected N-terminal _CH2 residues of a human IgM heavy chain, wherein the position of the cysteine in the N-terminal _CH1 that forms the interchain _CL - _CH1 disulfide bond is indicated (underlined); and

Residues in the upper hinge of IgG3 and IgG4, residues in the hinge of IgD, and residues in the C-terminal _CH1 and _CH2 of IgM where underlined residues indicate positions where one or more residues may be substituted with cysteine in an antibody of the invention.

Figure 2a shows the _CH1 cysteine residue (C127) that forms an interchain disulfide bond with a cysteine in the light chain and upper and core hinge residues of IgG1 wild type, IgG4 wild type, and positions where mutations are introduced into the IgG4 antibody of the present invention.

Figure 2b shows the _CH1 cysteine residue (C127) that forms an interchain disulfide bond with a cysteine in the light chain and the hinge residues of IgG3 wild type, as well as positions where one or more residues are substituted with cysteine in the IgG3 antibodies of the invention.

Figure 2c shows the _CH1 cysteine residue (C127) that forms an interchain disulfide bond with a cysteine in the light chain and selected _CH1 and _CH2 residues of IgM wild type, and positions where one or more residues are substituted with cysteine in the IgM antibodies of the invention.

Figure 2d shows the _CH1 cysteine residue (C128) that forms an interchain disulfide bond with a cysteine in the light chain and the hinge residues of IgD wild type, and positions where one or more residues are substituted with cysteine in the IgD antibodies of the invention.

Figure 3a shows mutations introduced into an IgG4 antibody according to the invention.

Figure 3b shows the positions of residues in the mutated heavy chain of the IgG4 antibody shown in Figure 3a and the predicted disulfide bonds that can form with a cysteine in the light chain (LC) or with another mutated heavy chain (HC), where a cysteine can bond to a cysteine in either the LC or the HC. The underlined chains are the predicted major disulfide bond arrangements.

Figure 4a shows mutations introduced into an IgG4 antibody according to the invention.

Figure 4b shows the positions of cysteine residues in the IgG4 antibody shown in Figure 4a and the predicted disulfide bonds that can be formed with cysteines in the light chain (LC) or heavy chain (HC). Where a cysteine can bond to a cysteine in the LC or HC, the underlined chains are the predicted major disulfide bond arrangements.

Figure 5 shows various sequences.

Figure 6 shows various sequences.

FIG7 shows Western blot analysis of antibodies according to the present invention, the upper gel shows the results using anti-human Fc antibody, and the bottom gel shows the results using anti-κ antibody.

Figure 8 shows Western blot analysis of antibodies according to the present invention, the top gel shows the results using anti-human Fc antibody, and the bottom gel shows the results using anti-human κ antibody.

Figure 9 shows Western blot analysis of antibodies according to the present invention, the top gel shows the results using anti-human Fc antibody, and the bottom gel shows the results using anti-human κ antibody.

Figure 10 shows Western blot analysis of antibodies according to the present invention, the top gel shows the results using anti-human Fc antibody, and the bottom gel shows the results using anti-human κ antibody.

FIG11 shows the results of Thermofluor analysis of the antibodies of the present invention, which show the thermal stability of the Fab and _CH2 domains.

FIG12 shows the results of Thermofluor analysis of the antibodies of the present invention, which show the thermal stability of the Fab and _CH2 domains.

FIG13 shows the results of Thermofluor analysis of the antibodies of the present invention, which show the thermal stability of the Fab and _CH2 domains.

FIG14 shows the results of Thermofluor analysis of the antibodies of the present invention, which show the thermal stability of the Fab and _CH2 domains.

Figure 15 shows the ranking of the thermal stability of selected antibodies of the invention.

Figure 16 shows heavy chain exchange at 16 hours under two GSH concentrations, where the first antibody is selected from IgG1 wild-type, IgG4 wild-type, and various mutant antibodies and the second antibody is IgG4 wild-type. The figure shows that the mutants have slightly less exchange than the wild-type IgG4 antibody and significantly more exchange than the IgG1 wild-type antibody and IgG4P antibody. This is advantageous because exchange can be used to prepare asymmetric antibodies of the present disclosure that have a lower susceptibility to undergo exchange in vivo than the wild-type IgG4 antibody. In some cases, increasing the concentration of a reducing agent, such as GSH, increases the amount of exchange observed.

Figure 17 shows asymmetric exchange analysis of mutants comprising type 1 variable regions having a substituted residue at position 241 and type 2 variable regions.

Figure 18 Asymmetric exchange analysis of IgG4 WT with type 1 variable regions and type 2 variable regions incubated with different S241 and core hinge cysteine mutants.

Figure 19 shows asymmetric exchange analysis of IgG4 S241P with type 1 variable regions and type 2 variable regions incubated with different S241 mutants, IgG4 C127S Y229C (Ab28).

FIG20 shows asymmetric exchange analysis of IgG4 C127S Y229C (No. 28) having type 1 variable region and type 2 variable region incubated with different S241 mutants and IgG4 WT.

Figure 21 shows asymmetric exchange analysis of double hinge mutants with type 1 variable regions and type 2 variable regions incubated with different mutants.

Sequence Overview

SEQ ID NO: 1 shows the _CH1 and hinge region sequences of an IgG1 wild-type antibody.

SEQ ID NO: 2 shows the _CH1 and hinge region sequences of an IgG4 wild-type antibody.

SEQ ID NO: 3 shows a portion of the constant region of a human wild-type kappa light chain.

SEQ ID NO: 4 shows a portion of the N-terminal sequence of the _CH1 domain of a human IgG1 antibody.

SEQ ID NO: 5 shows the hinge region of a human IgG1 antibody.

SEQ ID NO: 6 shows a portion of the N-terminal sequence of the _CH1 domain of a human IgG2 antibody.

SEQ ID NO: 7 shows the hinge region of a human IgG2 antibody.

SEQ ID NO: 8 shows a portion of the N-terminal sequence of the _CH1 domain of a human IgG3 antibody.

SEQ ID NO: 9 shows the hinge region of a human IgG3 antibody.

SEQ ID NO: 10 shows a portion of the N-terminal sequence of the _CH1 domain of a human IgG4 antibody.

SEQ ID NO: 11 shows the hinge region of human IgG4 antibody.

SEQ ID NOs: 12 to 37 show the _CH1 domain and hinge region sequences of antibodies 6, 7, 8, 15, 16, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 44, 45, 46, 47, 2, 3, 48, 28P, and 44P, respectively.

SEQ ID NOs: 38 to 63 show the _CH1 domain, hinge region, _CH2 domain, and CH3 domain sequences of antibodies 6, 7, 8, 15, 16, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 44, 45, 46, 47, 2, 3, 48, 28P, and 44P, respectively.

SEQ ID NO: 64 shows the wild-type IgG4 _CH2 and _CH3 domain sequences.

SEQ ID NO: 65 shows the wild-type IgG4 _CH2 and wild-type IgG1 _CH3 domain sequences.

SEQ ID NO: 66 shows the constant region sequence of human wild-type kappa light chain.

SEQ ID NO: 67 shows a portion of the N-terminal sequence of the _CH1 domain of a human IgD antibody.

SEQ ID NO: 68 shows a portion of the hinge region of a human IgG D antibody.

SEQ ID NO: 69 shows a portion of the N-terminal sequence of the _CH1 domain of a human IgM antibody.

SEQ ID NO: 70 shows a portion of the C-terminal sequence of the _CH1 domain of a human IgM antibody.

SEQ ID NO: 71 shows a portion of the _CH2 domain of a human IgM antibody.

SEQ ID NOs: 72 to 295 show the respective hinge regions.

SEQ ID NOs: 296 to 305 show the _CH1 domain and hinge region sequences of antibodies 1, 4, 5, 5P, 9, 10, 11, 12, 13, and 14, respectively.

SEQ ID NOs: 306 to 315 show the _CH1 domain, hinge region, _CH2 domain, and _CH3 domain sequences of antibodies 1, 4, 5, 5P, 9, 10, 11, 12, 13, and 14, respectively.

SEQ ID NOs: 316 to 322 show respective hinge sequences and portions thereof.

Detailed Description of the Invention

As used herein, "asymmetric antibodies" are antibodies in which the two heavy chains or fragments thereof have partially or completely different amino acid sequences in regions outside the variable regions, for example, having less than 98% similarity, such as less than 97, 96, or 95% similarity in the relevant regions. In one embodiment, there are 1, 2, 3, 4, or 5 amino acid differences or additions in a region of 10 consecutive amino acids. The amino acid sequences may also be of different lengths, which will inevitably result in differences in the amino acid sequences. Portions of the heavy chains may have similar or identical sequences, for example, the variable regions within the heavy chains may be identical or different.

In one embodiment, the heavy chain sequences of an antibody of the disclosure are covalently linked, e.g., by interchain disulfide bonds, such as those naturally present in the corresponding wild-type fragment or genetically engineered to exist at desired positions in the chain.

In one aspect, an antibody of the disclosure is characterized in that the first heavy chain IgG4 sequence or fragment has an IgG1 type hinge.

In one aspect, the antibodies of the present disclosure are characterized in that the heavy chain sequences or fragments all have an IgG1 type hinge.

The upper hinge and core hinge of wild-type IgG1 have the sequence EPKSCDKTHTCPPCP SEQ ID No:316.

The upper hinge and core hinge of wild-type IgG4 have the sequence ESKYGPPCPSCP SEQ ID No:317.

As used herein, an IgG1 type hinge is intended to mean one in which one or more, e.g. 1 to 5, e.g. 1, 2 or 3 amino acids are inserted into the IgG4 hinge, in particular between EPKYGPP SEQ ID No: 318 and CPSC, and/or one or more of the amino acids YGPP in the IgG4 hinge are replaced, e.g. by the corresponding amino acids in the IgG1 hinge, in particular G (from YGPP in the IgG4 hinge) is replaced by C or wherein Y (from YGPP in the IgG4 hinge) is replaced by C or S.

Thus, the present invention also provides an asymmetric hybrid antibody comprising a first IgG4 heavy chain having an upper hinge, a core hinge and a lower hinge, wherein the upper hinge and the core hinge in the or each heavy chain herein have a length of 13 to 17, e.g. 15, amino acids.

In one embodiment, the asymmetric hybrid antibody having a first IgG4 heavy chain has an upper hinge and a core hinge that are 15 amino acids in length.

In one embodiment, the upper hinge and the core hinge of at least the first heavy chain comprise the natural 12 amino acids found in an IgG4 hinge and an additional 3 amino acids, e.g., 3 alanine residues or 3 glycine residues or a combination thereof.

In one embodiment, the hinge has one of the following sequences:

In one embodiment, the upper hinge and the core hinge in at least the first IgG4 heavy chain of the present disclosure consist of the native IgG1 hinge, EPKSCDKTHTCPPC SEQ ID No: 96, or a derivative thereof, such as:

Typically the hinge regions in each heavy chain of an asymmetric hybrid antibody will at least be compatible, that is, when the heavy chains pair the arrangement will not be unstable, for example due to internal tension in the hinge region.

In one embodiment, the hinge region of each heavy chain comprises a sequence independently selected from the hinge sequences disclosed herein.

In one embodiment, the hinge of each heavy chain is similar or identical. This can be advantageous because it minimizes incompatibilities in the hinge regions of the two chains.

In other aspects, the present invention provides an asymmetric hybrid antibody comprising two IgG4 heavy chains, each comprising at least a variable region, a _CH1 domain and a hinge region, wherein in the first heavy chain:

the interchain cysteine at position 127 according to the Kabat numbering system in the _aCH1 domain is replaced with another amino acid; and

b. wherein the hinge in the or each heavy chain has a length of 12 to 17, for example 15, amino acids;

Part or all of the second heavy chain has an amino acid sequence different from that of the first heavy chain in at least the region outside the variable region.

Suitable hinges are described above.

In other aspects, the present invention also provides an asymmetric antibody comprising two IgG4 heavy chains, each comprising a _CH1 domain and a hinge region, wherein in the first heavy chain:

a. cysteine at position 127, numbered according to the Kabat numbering system, is replaced by another amino acid; and

b. cysteine at position 239 or cysteine at position 242 is substituted with another amino acid according to the Kabat numbering system;

Part or all of the second heavy chain has an amino acid sequence different from that of the first heavy chain in at least the region outside the variable region.

In one embodiment according to the latter aspect of the invention, at least the IgG4 heavy chain comprises 22 amino acids in the hinge, e.g. as described above.

Second heavy chain sequences and fragments include any antibody heavy chain, including IgGl, IgG2, IgG3, IgG4 (including the types described above), IgD, and IgM.

In other embodiments, the present invention also provides an asymmetric hybrid antibody comprising two IgG3 heavy chains, each comprising a _CH1 domain and a hinge region, for example, wherein in the first heavy chain:

The cysteine in the _aCH1 domain that forms an interchain disulfide bond with a cysteine in the light chain is replaced with another amino acid; and

b. one or more amino acids in the upper hinge region are substituted with cysteine;

Part or all of the second heavy chain has an amino acid sequence different from that of the first heavy chain in at least the region outside the variable region.

In other embodiments, the present invention provides an asymmetric hybrid antibody further comprising two IgM heavy chains each comprising a _CH1 domain and a _CH2 domain, e.g., wherein in the first heavy chain:

The cysteine in the _aCH1 domain that forms an interchain disulfide bond with a cysteine in the light chain is replaced with another amino acid; and

b. One or more amino acids in the _CH1 domain or _CH2 domain are substituted with cysteine;

Part or all of the second heavy chain has an amino acid sequence different from that of the first heavy chain in at least the region outside the variable region.

In other embodiments, the present invention provides an asymmetric hybrid antibody further comprising two IgD heavy chains, e.g., each of the heavy chains comprising a _CH1 domain and a hinge region, wherein in the first heavy chain:

The cysteine in the _aCH1 domain that forms an interchain disulfide bond with a cysteine in the light chain is replaced with another amino acid; and

b. one or more amino acids in the hinge region are substituted with cysteine;

Part or all of the second heavy chain has an amino acid sequence different from that of the first heavy chain in at least the region outside the variable region.

While not wishing to be bound by theory, it is speculated that the _CH3 region of IgG4 antibodies functions in the dynamic exchange process. Therefore, replacing the _CH3 domain of a non-IgG4 antibody with the _CH3 domain of an IgG4 antibody may render the mutant antibody more conducive to exchange.

In one embodiment, one or more cysteines that would naturally participate in the formation of an interchain disulfide bond between the light chain and the heavy chain are replaced by a non-cysteine amino acid as described in WO2005/003170 and WO2005/003171 (both of which are incorporated herein by reference).

In one embodiment, the human kappa light chain in the antibody or fragment according to the present disclosure has one or more substituted residues 171, 156, 202 or 203 as described in WO2008/038024 (which is incorporated herein by reference).

Those skilled in the art will appreciate that mutations generated for IgG4 antibodies can also be applied to other antibody isotypes or species having the same disulfide bond arrangement as IgG4 antibodies to provide improved antibodies. Specific examples of antibodies having the same disulfide bond arrangement as IgG4 antibodies are IgG3 antibodies, IgM antibodies, and IgD antibodies. As shown in Figure 1b, IgG3 and IgM have a cysteine at position 127 in the _CH1 domain, and IgD has a cysteine at position 128 in the _CH1 domain, which is equivalent to C127 in the _CH1 domain of IgG4 that forms an interchain disulfide bond with a cysteine in the light chain. Furthermore, as can be seen in Figure 1b, the upper hinge region of IgG3 and IgD, as well as the C-terminal region of the _CH1 domain and the N-terminal region of the _CH2 domain of IgM, do not contain a cysteine residue equivalent to the residue in the upper hinge region of IgG1. Therefore, the present invention also provides IgG3 antibodies, IgD antibodies, and IgM antibodies, in which the cysteine in the _CH1 domain that forms an interchain disulfide bond with the cysteine in the light chain is replaced by another amino acid, and in which one or more amino acids in a position structurally similar to the upper hinge region of IgG1 or IgG4 are replaced by cysteine. These mutated heavy chains can be used, for example, as the second heavy chain of an antibody according to the present disclosure.

In one embodiment, an antibody according to the present disclosure comprises two heavy chains of the IgG4 class.

In one embodiment, an antibody according to the present disclosure further comprises two light chains.

In one embodiment, an antibody according to the present disclosure comprises two variable regions.

In one embodiment, the two variable regions have the same specificity, that is, they are specific for the same antigen.

In one aspect, the present disclosure provides asymmetric hybrid antibodies in which the variable regions have different specificities, i.e., bispecific antibodies, that is, antibodies comprising two heavy chain variable regions, each of which is specific for a different antigen.

In one embodiment, each variable region can independently bind to a target antigen.

The present antibody format is advantageous because it can be readily obtained from conventional antibody production methods and utilizes naturally occurring machinery.

In one embodiment, the C-terminus of one or both heavy chains is fused to, for example, a domain antibody with specificity for a different antigen than that for which the variable region of the heavy chain is not specific.

Single variable domains (also referred to as single domain antibodies or dAbs) for use in the present invention can be produced using methods known in the art, and include those described in WO2005118642, Ward et al., 1989, Nature, 341, 544-546, and Holt et al., 2003, Trends in Biotechnology, 21, 484-490. In one embodiment, the single domain antibodies used in the present invention are heavy chain variable domains (VH) or light chain domains (VL). Each light chain domain can be of κ or λ subtype. Methods for separating VH and VL domains have been described in the art, see, for example, EP0368684 and Ward et al. (supra). Such domains can be derived from any appropriate species or antibody starting material. In one embodiment, single domain antibodies can be derived from rodents, humans, or other species. In one embodiment, single domain antibodies are humanized.

In one embodiment, the single domain antibody is derived from a phage display library using methods such as those described in WO 2005/118642, Jespers et al., 2004, Nature Biotechnology, 22, 1161-1165, and Holt et al., 2003, Trends in Biotechnology, 21, 484-490. Preferably, such single domain antibodies are fully human, but may also be derived from other species. It will be appreciated that the sequence of the single domain antibody may be modified after isolation to improve the properties of the single domain antibody, such as solubility, as described in Holt et al. (supra).

In one embodiment, the or each domain antibody is a VH or VHH.

In one embodiment, there are two domain antibodies, one fused to each heavy chain, wherein the two domain antibodies form a VH/VL pairing that bind to the antigen they are specific for cooperatively.

In one embodiment, the antibodies of the disclosure are isolated, that is, not located in the human or animal body.

Unless the context dictates otherwise, the terms "protein" and "polypeptide" are used interchangeably herein. "Peptide" is intended to mean 10 or fewer amino acids.

Unless otherwise indicated, the term "polynucleotide" includes genes, DNA, cDNA, RNA, mRNA, and the like.

As used herein, the term "comprising" in the context of this specification should be interpreted as "including".

In the context of the present invention, the term "wild-type" means an antibody that does not comprise any genetically engineered mutations as may occur naturally or may be isolated from the environment.

The nomenclature of substitution mutants herein consists of a letter followed by a number followed by a letter. The first letter represents the amino acid in the wild-type protein. The number refers to the amino acid position where the amino acid substitution occurs, and the second letter represents the amino acid that replaces the wild-type amino acid.

The residues in the variable and constant domains of antibodies are conventionally numbered according to the system devised by Kabat et al. This system is shown in Kabat et al., 1987, in Sequences of Proteins of Immunological Interest, US Department of Health and Human Services, NIH, USA (hereinafter "Kabat et al. (supra)").

Kabat residue nomenclature does not always correspond directly to the linear numbering of amino acid residues. The actual linear amino acid sequence may contain fewer amino acids than in the strict Kabat numbering or may contain additional amino acids corresponding to the shortening of, or insertion into, structural elements of the basic variable domain structure, whether framework or complementarity determining regions (CDRs). The correct Kabat numbering of residues for a given antibody can be determined by aligning residues with homology in the antibody sequence with sequences using the "standard" Kabat numbering.

Alternatively, the numbering of amino acid residues may be by the EU-index or EU numbering system (also described in Kabat et al., Sequences of Proteins of Immunological Interest, 5th ed., Public Health Service, National Institutes of Health, Bethesda, MD. (1991)).

Another numbering system for the amino acid residues of antibodies is the IMGT numbering system (Lefranc, M.-P. et al., Dev. Comp. Immunol., 29, 185-203 (2005)).

The Kabat numbering system is used in this specification, except where otherwise indicated where the EU numbering system or the IMGT numbering system is used.

The intrachain disulfide bond arrangement in the heavy and light chains is similar among the four IgG isotypes, however the interchain disulfide bond arrangement is unique to each isotype [reviewed by (Wypych, J., Li, M., Guo, A., Zhang, Z., Martinez, T., Allen, M.J., Fodor, S., Kelner, D.N., Flynn, G.C., Liu, Y.D., Bondarenko, P.V., Ricci, M.S., Dillon, T.M., Balland, A., 2008. Human IgG2 antibodies display disulphide-mediated structural isoforms. J Biol Chem. 283, 16194-16205)].

As shown in Figure 1b, the hinge region sequences of 4 kinds of IgG isotypes are different. Complete or hereditary hinge region is usually composed of residues 226 to 251 (numbered according to the Kabat numbering system). Figure 1b shows the upper section, core section and lower section of the hinge region of 4 kinds of IgG isotypes. For the IgG1 isotype, the upper hinge region is residues 226 to 238, the core hinge region is residues 239 to 243, and the lower hinge region is residues 244 to 251. For the IgG4 isotype, the upper hinge region is residues 226 to 238, the core hinge region is residues 239 to 243, and the lower hinge region is residues 244 to 251.

Thus, as shown in Figure 1a, the hinge in IgG1, comprising the upper hinge, core hinge, and lower hinge, is 23 amino acids in length. The upper hinge is 10 amino acids, the core hinge is 5 amino acids, and the lower hinge is 8 amino acids, see, for example, Figure 1b.

As shown in Figure 1a, the hinge in IgG4, which includes the upper hinge, core hinge, and lower hinge, is 20 amino acids in length. The upper hinge is 7 amino acids. The core hinge is 5 amino acids, and the lower hinge is 8 amino acids, see, for example, Figure 1b.

The novel mutant IgG4 antibodies according to the present invention have been developed by modifying the interchain disulfide bond arrangement within IgG4, in particular the CL- _CH1 interchain _disulfide bond between the light chain (LC) and the heavy chain (HC) has been modified.

Figure 1b shows human IgG heavy and light chain sequence segments of the IgG1-4 isotypes, with the cysteine positions (underlined) that form the _CL - _CH1 interchain disulfide bond indicated. The _CL - _CH1 inter-disulfide bond of IgG1 is formed just before the hinge region between LC C214 (Kabat numbering system) and C233 (Kabat numbering system) of the HC. In contrast, the _CH1 - _CL disulfide bond of IgG2, 3, and 4 is formed N-terminal to the intrachain disulfide bond of the HC between LC C214 and C127. The LC and HC sequences surrounding the cysteine residues involved in _CL - _CH1 disulfide bond formation are shown and aligned in Figure 1b.

The present inventors have investigated how the _CL - _CH1 disulfide bond affects the properties of IgG4 antibodies, including thermal stability, structural stability, disulfide isoform heterogeneity, affinity, and half-molecule exchange.

IgG4 mutants can be generated by replacing the cysteine residue at position 127 in _CH1 with another amino acid and replacing one or more amino acids in the upper hinge region with cysteine, preferably selected from the group consisting of amino acids at positions 227, 228, 229, and 230 of IgG4 according to the Kabat numbering system. Positions 227, 228, 229, or 230 are located at or near the equivalent structural position where cysteine 233 of IgG1 is located.

Each heavy chain may contain other mutations, including substitution of one or both of the cysteine residues 239 and 242 in the IgG4 hinge region with another amino acid. A mutation extending the IgG4 upper hinge region by 3 amino acids between positions 238 and 239 to the same length as the IgG1 hinge may also be included in some antibodies. An S241P mutation has also been introduced into some antibodies.

Thus, in one embodiment, an IgG4 antibody is provided in which cysteine 127 is replaced with another amino acid and the cysteine of the light chain is linked to the engineered cysteine at position 227, 228, 229 or 230 via a disulfide bond.

In one embodiment, the upper hinge region and the core hinge region are selected from one of the following sequences:

In one embodiment, the core hinge region in one or both heavy chain sequences or fragments thereof has the sequence CPPCP SEQ ID NO: 322.

While not wishing to be bound by theory, it is thought that this sequence may block the dynamic exchange of antibody arms at "in vivo" type concentrations, e.g., concentrations of reducing agent below 0.5 mM, particularly reducing agent concentrations of the order of 5 uM which are considered physiologically relevant (Zilmer et al., 2005 Drug Design Reviews vol. 2, no. 2, pp. 121-127, 2005).

Mutations of the antibodies of the present invention will now be described in more detail. Methods for replacing amino acids are well known in the art of molecular biology. Such methods include, for example, site-directed mutagenesis or de novo design of synthetic sequences using methods such as PCR to delete and/or replace amino acids.

Figure 2a shows the hinge residues of IgG1 wild type, IgG4 wild type and the positions where mutations are introduced into the antibodies of the present invention. Numbering is based on the Kabat numbering system.

The antibodies according to the present invention comprise a mutation at position 127 (C127) in which the cysteine residue is replaced by another amino acid, preferably an amino acid that does not contain a sulfhydryl group. By substitution or replacement, we mean that in the place where interchain cysteine 127 is normally found in the heavy chain of an antibody, another amino acid is replaced. The mutation at C127 can be any suitable mutation of 1, 2 or 3 nucleotides of the nucleotide encoding the amino acid at position 127, which changes the amino acid residue from cysteine to another suitable amino acid. Examples of suitable amino acids include serine, threonine, alanine, glycine or any polar amino acid. Particularly preferred amino acids are serine.

Another amino acid substitution for cysteine at position 127 removes the cysteine in the _CH1 domain that normally forms a disulfide bond with a cysteine in the light chain of wild-type IgG4. Therefore, in order for the light chain to pair with the heavy chain via interchain disulfide bond formation, the light chain must form a disulfide bond with a cysteine located in the hinge region of the heavy chain.

In a first aspect of the invention, an antibody according to the invention comprises a heavy chain in which one or more of the amino acids at positions selected from 227, 228, 229 and 230 (numbered according to the Kabat numbering system) are substituted with cysteine. Thus, an antibody according to the invention may have one or more of the following mutations: S227C, K228C, Y229C, G230C.

Preferably only one residue selected from 227, 228, 229 and 230 is substituted with a cysteine residue.

Particularly preferred antibodies of the invention have the mutation Y229C or G230C.

The inclusion of a cysteine residue at a position selected from 227, 228, 229 and 230 in the hinge region of the heavy chain provides a new position for interchain disulfide bond formation between the heavy and light chains.

Other mutations can be introduced into the antibody of this aspect of the present invention. In one embodiment, with another amino acid, preferably the cysteine (C239) on the position 239 and/or the cysteine (C242) on the position 242 of the amino acid replacement heavy chain according to the Kabat numbering system numbering that does not comprise sulfhydryl. About replacement or displacement, we mean the position that cysteine 239 and/or cysteine 242 are usually found in the antibody heavy chain, and another amino acid is in its position. The mutation on C239 and/or C242 can be any suitable mutation of 1,2 or 3 nucleotides of the nucleotides encoding amino acids, and the mutation changes the amino acid residue from cysteine to another suitable amino acid. The example of suitable amino acid includes serine, threonine, alanine, glycine or any polar amino acid. Particularly preferred amino acid is serine.

In one embodiment, the cysteine at position 239 of the heavy chain is replaced by another amino acid, and the cysteine at position 242 of the heavy chain is replaced by another amino acid. In this embodiment, the replacement of C239 and C242 removes two cysteine residues in the hinge region of the heavy chain that normally form inter-heavy chain disulfide bonds with corresponding cysteines in the other heavy chain. The resulting half-molecule can form a complete antibody molecule by non-covalent bonding between the two heavy chains.

In an alternative embodiment, the cysteine at position 239 of the heavy chain is substituted with another amino acid. In this embodiment, the cysteine at position 242 is not substituted with another amino acid.

In another alternative embodiment, the cysteine at position 242 of the heavy chain is replaced by another amino acid. In this embodiment, the cysteine at position 239 is not replaced by another amino acid.

Substitution of C239 or C242 leaves a cysteine in the heavy chain that is capable of forming an inter-heavy chain disulfide bond with a cysteine in another heavy chain. Without wishing to be bound by theory, it is believed that substitution of a cysteine in the hinge region, particularly substitution of C239, reduces the formation of intra-chain disulfide bonds in the hinge region, thereby reducing the formation of half-antibody molecules.

In one embodiment of the present invention, the proline at position 240 may be substituted with another amino acid.

In one embodiment of the present invention, the serine at position 241 may be substituted with another amino acid.

In one embodiment of the invention in which the serine at position 227 is substituted with cysteine, the antibody preferably does not contain mutations at positions C239 and C242. In another embodiment in which the serine at position 227 is substituted with cysteine, the cysteine at position 239 of the heavy chain is preferably substituted with another amino acid but the cysteine at position 242 is not substituted with another amino acid.

In one embodiment, the antibodies of the present invention include an IgG4 heavy chain in which one or more amino acids are inserted between amino acids 226 and 243 through mutation. The number of amino acids inserted can be 1 to 10, 1 to 5, 1 to 3, preferably 1, 2, 3 or 4 amino acids are inserted. Preferably, amino acids are inserted between amino acids 238 and 239. Any suitable amino acid, such as alanine, glycine, serine or threonine and combinations thereof, can be inserted in the hinge region. Preferably, 3 alanines (AAA), 3 glycines (GGG), 3 serines (SSS) or 3 threonines (TTT) or threonine, histidine and another threonine (THT) are inserted. It is believed that antibodies of the present invention comprising an IgG4 heavy chain in which 3 amino acids are inserted through mutation in the hinge region show enhanced stability, such as thermal stability.

Another mutation that can be introduced into the antibodies according to the present invention is the S241P mutation. This mutation has previously been shown to reduce the formation of half molecules at biologically relevant concentrations (Angal, S. et al., 1993. A single amino acid substitution abolishes the heterogeneity of chimeric mouse/human (IgG4) antibody. Mol Immunol 30, 105-108). Surprisingly, it has been found that mutant antibodies of the present invention comprising the S241P mutation exhibit some heavy chain exchange in vitro under strongly reducing conditions compared to IgG4P (IgG4 with S241P). This allows the in vitro generation of bispecific antibodies from the mutant IgG4 antibodies of the present invention.

The antibodies according to the present invention may comprise one or more other mutations in the hinge region. For example, the antibody may further comprise one or more of the following mutations: S227P, Y229S, P237D and P238K.

In one embodiment, an antibody according to the invention effectively comprises an IgG1 hinge region (upper hinge and core hinge) from residues 226 to 243. Thus, an antibody according to the invention comprises a hinge region in which the glycine at position 230 is substituted with a cysteine, the serine at position 227 is substituted with a proline, the tyrosine at position 229 is substituted with a serine, the proline at position 237 is substituted with an aspartic acid, the proline at position 238 is substituted with a lysine, the amino acid sequence threonine-histidine-threonine is inserted between positions 238 and 239, and the serine at position 241 is substituted with a proline. These mutations may also be written as S227P, Y229S, G230C, P237D, P238KTHT, and S241P, as shown in FIG2 a.

The antibodies according to the present invention preferably have an IgG4 lower hinge (APEFLGGP SEQ ID No: 321) from residues 244 to 251. Without wishing to be bound by theory, it is believed that the IgG4 lower hinge region contributes to the loss of effector function of IgG4 antibodies.

In a second aspect of the invention, the asymmetric hybrid antibody of the present disclosure comprises a heavy chain in which the cysteine at position 127 is substituted with another amino acid, as described above, and the cysteine at position 239 or position 242 (numbered according to the Kabat numbering system) in the heavy chain is substituted with another amino acid. In this second aspect, none of the residues at positions 227, 228, 229, and 230 are substituted with a cysteine residue. Thus, there is provided:

An asymmetric hybrid antibody comprising two heavy chains each comprising at least a variable region, a hinge region and a _CH1 domain, wherein the first heavy chain or a fragment thereof is characterized in that it is of class IgG4 and has:

a. the interchain cysteine at position 127 numbered according to the Kabat numbering system is replaced by another amino acid; and

b. optionally, the cysteine at position 239 or position 242 is substituted with another amino acid according to the Kabat numbering system;

The second heavy chain or fragment thereof is characterized in that it has an amino acid sequence that is different from that of the first heavy chain in regions outside the variable region.

In a second aspect of the invention, the antibody may comprise one or more additional mutations. In one embodiment, the antibody comprises at least a first IgG4 heavy chain mutated to insert three amino acids between amino acids 226 and 243, preferably between amino acids 238 and 239, as described above. In other embodiments, the antibody comprises the mutation S241P. In other embodiments, the antibody may further comprise one or more of the following mutations: S227P, Y229S, P237D, and P238K.

In one embodiment, the invention provides an asymmetric hybrid antibody comprising two heavy chains (e.g., two heavy chains independently) each comprising a variable region, a _CH1 domain, and a hinge, and each heavy chain independently comprises a sequence selected from one of the following sequences: SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 19, SEQ ID NO: 20, SEQ ID NO: 21, SEQ ID NO: 22, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 25, SEQ ID NO: 26, SEQ ID NO: 27, SEQ ID NO: 28, SEQ ID NO: 29, SEQ ID NO: 30, SEQ ID NO: 31, SEQ ID NO: 32, SEQ ID NO: 33, SEQ ID NO: 34, SEQ ID NO: 35, SEQ ID NO: 36, and SEQ ID NO: 37.

In one embodiment, the asymmetric hybrid antibody of the invention comprises two heavy chains, each heavy chain comprising a variable region, a _CH1 domain and a hinge (e.g., both heavy chains independently), and each heavy chain independently comprises a sequence selected from one of the following sequences: SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 19, SEQ ID NO: 20, SEQ ID NO: 21, SEQ ID NO: 22, SEQ ID NO: 23, SEQ ID NO: 24, SEQ ID NO: 25, SEQ ID NO: 26, SEQ ID NO: 29, SEQ ID NO: 30, SEQ ID NO: 31, SEQ ID NO: 32, SEQ ID NO: 35, SEQ ID NO: 36, and SEQ ID NO: 37. More specifically, the asymmetric hybrid antibody of the present invention comprises two heavy chains, each heavy chain comprising a variable region, a _CH1 domain and a hinge (e.g., both heavy chains independently), and each heavy chain independently comprises a sequence selected from one of the following sequences: SEQ ID NO: 12, SEQ ID NO: 13, SEQ ID NO: 14, SEQ ID NO: 15, SEQ ID NO: 16, SEQ ID NO: 17, SEQ ID NO: 18, SEQ ID NO: 19, SEQ ID NO: 20, SEQ ID NO: 29, SEQ ID NO: 30, SEQ ID NO: 31, SEQ ID NO: 32, SEQ ID NO: 35, SEQ ID NO: 36, and SEQ ID NO: 37.

Thus, the present invention provides two heavy chains of an asymmetric hybrid antibody, each of said heavy chains comprising a variable domain, a _CH1 domain and a hinge, a _CH2 domain and a _CH3 domain (e.g., both heavy chains independently), and each heavy chain comprising a sequence independently selected from one of the following sequences: SEQ ID NO: 38, SEQ ID NO: 39, SEQ ID NO: 40, SEQ ID NO: 41, SEQ ID NO: 42, SEQ ID NO: 43, SEQ ID NO: 44, SEQ ID NO: 45, SEQ ID NO: 46, SEQ ID NO: 47, SEQ ID NO: 48, SEQ ID NO: 49, SEQ ID NO: 50, SEQ ID NO: 51, SEQ ID NO: 52, SEQ ID NO: 53, SEQ ID NO: 54, SEQ ID NO: 55, SEQ ID NO: 56, SEQ ID NO: 57, SEQ ID NO: 58, SEQ ID NO: 59, SEQ ID NO: 60, SEQ ID NO: 61, SEQ ID NO: 62, SEQ ID NO: 63, SEQ ID NO: 64, SEQ ID NO: 65 NO:62 and SEQ ID NO:63.

In one embodiment, the asymmetric hybrid antibody of the invention comprises two heavy chains, each heavy chain comprising a variable region, a _CH1 domain, and a hinge region, wherein each heavy chain independently comprises SEQ ID NO: 36 (antibody 28P), SEQ ID NO: 37 (antibody 44P), or SEQ ID NO: 35 (antibody 48).

In one embodiment, the asymmetric hybrid antibody of the invention comprises two heavy chains, each heavy chain comprising a variable region, a _CH1 domain, a hinge region, a _CH2 domain, and a _CH3 domain, wherein each heavy chain independently comprises SEQ ID NO: 62 (antibody 28P), SEQ ID NO: 63 (antibody 44P), or SEQ ID NO: 61 (antibody 48).

In any of the above embodiments, the second heavy chain of the antibody can be selected from any heavy chain sequence disclosed herein.

Table 1 below lists exemplary antibodies having mutations that have been introduced compared to the IgG4 wild-type sequence. Table 1 also includes wild-type IgG1 and IgG4 antibodies and a control antibody.

Table 1

In one embodiment, a first IgG4 heavy chain sequence, each of which comprises the _CH1 and upper hinge mutations and the core hinge sequence as described in Table 2, is combined with a second IgG4 heavy chain sequence:

Table 2

Thus, the present invention provides an asymmetric hybrid antibody comprising two heavy chains, each of which comprises at least a variable region, a hinge region and a _CH1 domain, wherein the first heavy chain and the second heavy chain sequence are IgG4 heavy chain sequences selected from the combination of the first and second heavy chain sequence mutations listed in Table 2.

The asymmetric hybrid antibody may comprise a first and a second heavy chain, wherein each heavy chain constant sequence comprises mutations in the _CH1 domain and hinge region as described above, and wherein the mutations in the _CH1 domain and hinge region in each heavy chain are different. Alternatively, the first heavy chain constant sequence may comprise mutations in the _CH1 domain and hinge region as described above and the second heavy chain constant sequence is IgG4 wild type or IgG4 wild type with an S241P mutation.

In one embodiment of the invention, the asymmetric hybrid antibody is a bispecific antibody, wherein each heavy chain has a different variable region. The antibody preferably further comprises two light chains, wherein each heavy chain-light chain pair (Fab) has a different variable region.

As used herein, "different variable regions" are intended to mean that the variable regions have specificity for different antigens. That is, the antigen to which each variable region is specific is a different antigen or a different part of an antigen, such as a different epitope.

As used herein, "specific" refers to the fact that a binding domain recognizes a target antigen with greater affinity and/or avidity (e.g., 10, 20, 50, 10, or 1000-fold) than other antigens to which it is non-specifically directed. This does not necessarily mean that the specific binding region does not bind to any non-target antigens, but rather that the interaction with the target allows it to be used to purify the target antigen (for which it is specific) from a complex mixture of antigens (including antigens in the same protein family).

In one embodiment, the antibodies according to the present disclosure are isolated.

As used herein, isolated is intended to refer to antibodies that are separated from the human body, for example, produced by recombinant techniques, purified using techniques such as chromatography, and/or in a pharmaceutical formulation.

As used herein, the term "antibody" includes intact (whole) antibodies and functionally active fragments comprising two heavy chains, each comprising a _VH domain, a _CH1 domain, and a hinge region. Antibodies according to the present invention preferably comprise at least one light chain. Thus, the term "antibody" in the present invention encompasses bivalent, trivalent, or tetravalent antibodies, Fab' dimers, and F(ab') ₂ fragments, as well as intact antibody molecules comprising two light chains paired with a heavy chain.

As is well known in the art, a typical Fab' molecule comprises a heavy and light chain pair, wherein the heavy chain comprises a variable region _VH , a constant domain _CH1 and a hinge region, and the light chain comprises a variable region _VL and a constant domain _CL .

In one embodiment, a dimer of a Fab' according to the present disclosure is provided, e.g., dimerization may occur via a hinge.

In one embodiment, the heavy chain comprises a _CH2 domain and a _CH3 domain and optionally a _CH4 domain. In one embodiment, the antibody comprises two heavy chains, each of which is as defined above in the first or second aspect of the invention. The antibody according to the invention also preferably comprises two light chains, which may be identical or different. In embodiments of the invention in which a bispecific antibody comprising two heavy chains and two light chains as defined above is provided, the two light chains have different variable regions and may have identical or different constant regions.

In one embodiment, the _CH2 and _CH3 domains used can be mutated, for example, to reduce aggregate formation of IgG4 antibodies. US 2008/0063635 Takahashi et al. have studied mutants of IgG4 in which the arginine at position 409 (409 according to the EU numbering system or 440 according to the Kabat numbering system) in the _CH3 domain is replaced by lysine, threonine, methionine, or leucine to inhibit aggregate formation at low pH. Other mutations at L235, D265, D270, K322, P329, and P331 (L235, D265, D270, K322, P329, and P331 according to the EU numbering system or L248, D278, D283, K341, P348, and P350 according to the Kabat numbering system) are also taught to reduce CDC activity. WO2008/145142 Van de Winkel et al. discloses stable IgG4 antibodies having a reduced ability to undergo "Fab-arm exchange" by substitution of an arginine residue at position 409, a Phe residue at position 405 or a Lys at position 370 (R409, F405 and K370 according to the EU numbering system or R440, F436 and K393 according to the numbering system of Kabat), even in the absence of an S228P (S228 according to the EU numbering system or S241 according to the Kabat numbering system) mutation in the hinge region.

In one embodiment, an antibody of the invention is a complete asymmetric mixed antibody comprising two light chains and two heavy chains, wherein each heavy chain comprises an IgG4 _CH1 in which the cysteine at position 127, numbered according to the Kabat numbering system, is substituted with another amino acid, an IgG1 upper and middle hinge region, an IgG4 lower hinge region, a _CH2 domain, and a _CH3 domain.

The complete hinge region of an IgG4 antibody typically consists of residues 226 to 251 (based on numbering according to the Kabat numbering system). However, the hinge region can be shortened or extended as needed. For example, in the antibody according to the first aspect of the invention, the wild-type amino acid is replaced by a cysteine residue at position 227, 228, 229, or 230, and the hinge region can terminate after the new cysteine residue at position 227, 228, 229, or 230. The antibody according to the present invention may also comprise one or more other amino acids at the N-terminus and/or C-terminus of the hinge region. In addition, other features of the hinge can be controlled, such as the distance between the hinge cysteine and the light chain interchain cysteine, the distance between the cysteines of the hinge, and the composition of other amino acids in the hinge that can affect the properties of the hinge, such as flexibility. For example, glycine can be incorporated into the hinge to increase rotational flexibility or proline can be incorporated to reduce flexibility. Alternatively, a combination of charged or hydrophobic residues can be incorporated into the hinge to confer multimerization or purification properties. Other modified hinge regions can be entirely synthetic and can be designed to have desired properties such as length, composition, and flexibility.

When present, the constant region domains used in the present invention (particularly in the Fc domain) are preferably of the IgG4 isotype where antibody effector functions are not required. Accordingly, each heavy chain preferably comprises an IgG4 _CH2 domain and a _CH3 domain, as shown in SEQ ID NO: 64.

It will be appreciated that sequence variants of the Fc constant region domains may also be used.

In one embodiment, each heavy chain comprises an IgG4 _CH2 and _CH3 domain in which the arginine at position 409 (EU numbering) is substituted with lysine, threonine, methionine, or leucine to inhibit aggregate formation at low pH (US 2008/0063635 Takahashi et al.). Mutations at L235, D265, D270, K322, P331, and P329 (numbered according to the EU numbering system) are also taught to reduce CDC activity (US 2008/0063635 Takahashi et al.).

Each heavy chain may comprise the mutations taught in WO 2008/145142 Van de Winkel et al., which discloses stable IgG4 antibodies with reduced ability to undergo Fab-arm exchange by substitution with an arginine residue at position 409, a Phe residue at position 405, or a Lys at position 370 (numbered according to the EU numbering system).

In one embodiment, each heavy chain comprises an IgG4 _CH2 domain and an IgG1 _CH3 domain, as shown in SEQ ID NO:65.

In embodiments of the invention in which the antibody is a mutant IgG3, IgD, or IgM antibody, each heavy chain preferably comprises a _CH2 domain and a _CH3 domain, and optionally a _CH4 domain. In an IgG3 antibody, each heavy chain preferably comprises an IgG3 _CH2 domain and an IgG3 _CH3 domain. In an IgD antibody, each heavy chain preferably comprises an IgD _CH2 domain and an IgD _CH3 domain. In an IgM antibody, each heavy chain preferably comprises an IgM _CH2 domain, an IgM _CH3 domain, and an IgM _CH4 domain.

In one embodiment, the antibody is a monoclonal, fully human, humanized or chimeric antibody fragment. In one embodiment, the antibody is fully human or humanized.

Monoclonal antibodies can be prepared by any method known in the art, such as the hybridoma technique (Kohler & Milstein, Nature, 1975, 256, 495-497), the trioma technique, the human B-cell hybridoma technique (Kozbor et al., Immunology Today, 1983, 4, 72), and the EBV-hybridoma technique (Cole et al., "Monoclonal Antibodies and Cancer Therapy", pp. 77-96, Alan R. Liss, Inc., 1985).

Antibodies for use in the present invention can also be produced using the single lymphocyte antibody method by cloning and expressing immunoglobulin variable region cDNAs produced from a single lymphocyte selected for the production of a specific antibody by the methods described, for example, by Babcook, J. et al., Proc. Natl. Acad. Sci. USA, 1996, 93(15), 7843-7848, WO 92/02551, WO 2004/051268, and WO 2004/106377.

Humanized antibodies are antibody molecules from a non-human species having one or more complementarity determining regions (CDRs) from the non-human species and framework regions from a human immunoglobulin molecule, optionally including one or more donor residues from the non-human species (see, e.g., US 5,585,089).

Antibodies for use in the present invention can also be produced using various phage display methods known in the art, including those described by Brinkman et al., J. Immunol. Methods, 1995, 182, 41-50; Ames et al., J. Immunol. Methods, 1995, 184, 177-186; Kettleborough et al., Eur. J. Immunol., 1994, 24, 952-958; Persic et al., Gene, 1997 187, 9-18; and Burton et al., Advances in Immunology, 1994, 57, 191-280; WO 90/02809; WO 91/10737; WO 92/01047; WO 92/18619; WO 93/11236; WO 95/15982; WO 95/20401; US 5,698,426; 5,223,409; 5,403,484; 5,580,717; 5,427,908; 5,750,753; 5,821,047; 5,571,698; 5,427,908; 5,516,637; 5,780,225; 5,658,727; 5,733,743 and 5,969, 108. Similarly, transgenic mice or other organisms, including other mammals, can be used to generate humanized antibodies.

Fully human antibodies are those wherein the variable region and constant region (when present) of heavy and light chains are all derived from people, or are substantially identical to sequences in people's origin but not necessarily from those antibodies of identical antibodies. The example of a fully human antibody may include, for example, antibodies produced by the above-mentioned phage display method and antibodies produced by mice, in which mouse immunoglobulin variable regions and/or constant region genes have been replaced by their human counterparts, as generally described in EP0546073B1, US 5,545,806, US 5,569,825, US 5,625,126, US 5,633,425, US 5,661,016, US5,770,429, EP 0438474B1 and EP0463151B1.

The antibody starting material for the present invention can be prepared by using recombinant DNA technology, which includes the manipulation and re-expression of DNA encoding antibody variable and constant regions. Standard molecular biology techniques can be used to modify, add or delete amino acids or domains as needed. Any changes to the variable or constant regions are still included in the terms 'variable' and 'constant' regions used in this article.

The antibody starting material can be obtained from any species including, for example, mice, rats, rabbits, hamsters, camels, llamas, goats or humans. The part of the antibody can be obtained from more than one species, for example, the antibody can be chimeric. In one example, the constant region is from one species and the variable region is from another species. The antibody starting material can also be modified. In another example, recombinant DNA engineering technology has been used to produce the variable region of the antibody. Such engineered forms include, for example, those produced from natural antibody variable regions by the amino acid sequence of natural antibodies or by insertion, deletion or change thereof. The specific examples of this type include those comprising at least one CDR from a kind of antibody and optionally one or more framework region amino acids and the remainder of the variable region domain is from the engineered variable region domain of a second antibody. Methods for generating and making these antibodies are well known in the art (see, e.g., Boss et al., U.S. Pat. No. 4,816,397; Cabilly et al., U.S. Pat. No. 6,331,415; Shrader et al., WO 92/02551; Ward et al., 1989, Nature, 341,544; Orlandi et al., 1989, Proc. Natl. Acad. Sci. USA, 86, 3833; Riechmann et al., 1988, Nature, 322,323; Bird et al., 1988, Science, 242,423; Queen et al., U.S. Pat. No. 5,585,089; Adair, WO 91/09967; Mountain and Adair, 1992, Biotechnol. Genet. Eng. Rev, 10, 1-142; Verma et al., 1998, Journal of Immunological Biology). Methods, 216, 165-181).

In one embodiment, the antibody comprises a pair of variable domains that form a binding domain, which are cognate pairs. As used herein, a cognate pair is intended to refer to a naturally occurring variable domain pair, that is, one that is isolated from a single antibody or antibody-expressing cell.

The variable domains may have been optimized and/or humanized.

Optimized/humanized variable domains derived from a cognate pair are still considered a cognate pair after optimization/humanization.

The invention therefore extends to human, humanized or chimeric molecules.

In one embodiment, the molecule specifically binds to the target antigen. As used herein, specific binding is intended to refer to a molecule that has a high affinity for the target antigen (for which the molecule is specific) and binds with a low or much lower affinity (or not at all) to an antigen for which the molecule is not specific. Methods of measuring affinity are known to those skilled in the art and include assays such as BIAcore ^™ .

The antibody molecules of the present invention suitably have high binding affinity (particularly nanomolar or picomolar). Affinity can be measured using any suitable method known in the art, including BIAcore ^™ . In one embodiment, the molecules of the present invention have a binding affinity of about 100 pM or better. In one embodiment, the molecules of the present invention have a binding affinity of about 50 pM or better. In one embodiment, the molecules of the present invention have a binding affinity of about 40 pM or better. In one embodiment, the molecules of the present invention have a binding affinity of about 30 pM or better. In one embodiment, the molecules of the present invention are fully human or humanized and have a binding affinity of about 100 pM or better.

As used herein, a derivative of a naturally occurring domain is intended to mean one in which 1, 2, 3, 4 or 5 amino acids have been substituted or deleted in the naturally occurring sequence, e.g., to optimize the properties of the domain, e.g., by eliminating undesirable properties (but wherein the characteristic properties of the domain are retained).

In one embodiment, the antibody molecule of the invention comprises one or more albumin binding peptides. In vivo the peptides bind to albumin which increases the half-life of the molecule.

The albumin binding peptide may be added from one or more of the variable regions, the hinge or the C-terminus of the molecule, or anywhere else that does not interfere with the antigen binding properties of the molecule.

Examples of albumin binding peptides are provided in WO 2007/106120.

It will be appreciated by those skilled in the art that antibodies may undergo a variety of post-translational modifications. The type and extent of these modifications generally depend on the host cell line used to express the molecule and the culture conditions. Such modifications may include changes in glycosylation, methionine oxidation, diketopiperazine formation, aspartic acid isomerization, and asparagine deamidation. Common modifications are the loss of carboxy-terminal basic residues (e.g., lysine or arginine) due to the action of carboxypeptidases (as described in Harris, RJ. Journal of Chromatography 705: 129-134, 1995).

If necessary, the molecules used in the present invention can be conjugated to one or more effector molecules. It should be understood that the effector molecule may comprise a single effector molecule or two or more such molecules (so connected to form a single part that can be connected to the antibody molecule of the present invention). When it is desired to obtain an antibody according to the present invention that is connected to an effector molecule, this can be prepared by standard chemical or recombinant DNA methods (wherein the antibody is directly or through a coupling agent connected to the effector molecule). The technology for conjugating such effector molecules to antibodies is well known in the art (see, Hellstrom et al., Controlled Drug Delivery, 2nd Ed., Robinson et al., eds., 1987, pp. 623-53; Thorpe et al., 1982, Immunol. Rev., 62: 119-58 and Dubowchik et al., 1999, Pharmacology and Therapeutics, 83, 67-123). Specific chemical methods include, for example, those described in WO 93/06231, WO 92/22583, WO 89/00195, WO 89/01476 and WO 03031581. Alternatively, when the effector molecule is a protein or polypeptide, linkage may be achieved using recombinant DNA methods, for example those described in WO 86/01533 and EP 0392745.

As used herein, the term effector molecule includes, for example, anti-tumor agents, drugs, toxins, biologically active proteins such as enzymes, other antibodies or antibody fragments, synthetic or naturally occurring polymers, nucleic acids and fragments thereof, such as DNA, RNA and fragments thereof, radionuclides, in particular radioiodides, radioisotopes, chelated metals, nanoparticles and reporter groups, such as fluorescent compounds or compounds detectable by NMR or ESR spectroscopy.

Examples of effector molecules include cytotoxins or cytotoxic agents, including any agent that is detrimental to (e.g., kills) cells. Examples include combrestatin, dolastatin, epothilone, staurosporine, maytansinoid, spongistatin, rhizoctonia, halichondrin, bacillus thuringiensis, hemiesterlins, taxol, cytochalasin B, gramicidin D, ethidium bromide, emetine, mitomycin, etoposide, tenoposide, vincristine, vinblastine, colchicine, doxorubicin, daunorubicin, dihydroxy anthracin dione, mitoxantrone, mithramycin, actinomycin D, 1-dehydrotestosterone, glucocorticoids, procaine, tetracaine, lidocaine, propranolol, and puromycin, and analogs or homologs thereof.

Effector molecules also include, but are not limited to, antimetabolites (e.g., methotrexate, 6-mercaptopurine, 6-thioguanine, cytarabine, 5-fluorouracil, decarbazine), alkylating agents (e.g., mechlorethamine, thioepachlorambucil, melphalan, carmustine (BSNU) and lomustine (CCNU), cyclothosphamide, busulfan, dibromomannitol, streptozotocin, mitomycin C, and cis-dichlorodiamineplatinum (II) (DDP) cisplatin), anthracyclines (e.g., daunorubicin (formerly daunomycin) and doxorubicin), antibiotics (e.g., dactinomycin (formerly actinomycin), bleomycin, mithramycin, anthramycin (AMC), calicheamicin, or duocarmycin), and antimitotic agents (e.g., vincristine and vinblastine).

Other effector molecules may include chelated radionuclides such as ¹¹¹ In and ⁹⁰ Y, Lu ¹⁷⁷ , Bismuth ²¹³ , Californium ²⁵² , Iridium ¹⁹² , and Tungsten ¹⁸⁸ /Rhenium ¹⁸⁸ ; or drugs such as, but not limited to, alkylphosphocholines, topoisomerase I inhibitors, taxanes, and suramin.

Other effector molecules include proteins, peptides, and enzymes. Target enzymes include, but are not limited to, proteases, hydrolases, lyases, isomerases, and transferases. Target proteins, polypeptides, and peptides include, but are not limited to, immunoglobulins, toxins such as abrin, ricin A, Pseudomonas exotoxin, or diphtheria toxin, proteins such as insulin, tumor necrosis factor, α-interferon, β-interferon, nerve growth factor, platelet-derived growth factor, or tissue-type plasminogen activator, thrombotic agents, or anti-angiogenic agents such as angiostatin or endostatin, or biological response modifiers such as lymphokines, interleukin-1 (IL-1), interleukin-2 (IL-2), granulocyte-macrophage colony stimulating factor (GM-CSF), granulocyte colony stimulating factor (G-CSF), nerve growth factor (NGF), or other growth factors and immunoglobulins.

Other effector molecules may include detectable substances for diagnosis, for example. Examples of detectable substances include various enzymes, prosthetic groups, fluorescent materials, luminescent materials, bioluminescent materials, radionuclides, positron-emitting metals (for positron emission tomography), and non-radioactive paramagnetic metal ions. For metal ions that can be conjugated to antibodies for use as diagnostic agents, see generally U.S. Patent No. 4,741,900. Suitable enzymes include horseradish peroxidase, alkaline phosphatase, β-galactosidase, or acetylcholinesterase; suitable prosthetic groups include streptavidin, avidin, and biotin; suitable fluorescent materials include umbelliferone, fluorescein, fluorescein isothiocyanate, rhodamine, dichlorotriazinylamine fluorescein, dansyl chloride, and phycoerythrin; suitable luminescent materials include luminol; suitable bioluminescent materials include luciferase, luciferin, and aequorin; and suitable radionuclides include ¹²⁵ I, ¹³¹ I, ¹¹¹ In, and ⁹⁹ Tc.

In another example, the effector molecule can increase the in vivo half-life of the antibody and/or reduce the immunogenicity of the antibody and/or enhance the delivery of the antibody across the epithelial barrier to the immune system. Examples of suitable effector molecules of this type include polymers, albumin, albumin binding proteins or albumin binding compounds such as those described in WO 05/117984.

When the effector molecule is a polymer, it may typically be a synthetic or naturally occurring polymer, such as an optionally substituted linear or branched polyalkylene, polyalkenylene or polyoxyalkylene polymer or a branched or linear polysaccharide, such as a homo- or hetero-polypeptide.

Specific optional substituents that may be present on the above-mentioned synthetic polymers include one or more hydroxyl, methyl or methoxy groups.

Specific examples of synthetic polymers include optionally substituted linear or branched poly(ethylene glycol), poly(propylene glycol), poly(vinyl alcohol) or derivatives thereof, particularly optionally substituted poly(ethylene glycol) such as methoxypoly(ethylene glycol) or derivatives thereof.

Specific naturally occurring polymers include lactose, amylose, dextran, glycogen or derivatives thereof.

As used herein, "derivatives" are intended to include reactive derivatives, such as sulfhydryl selective reactive groups such as maleimides, etc. The reactive group can be attached to the polymer directly or through a linker fragment. It should be understood that the residue of such a group will in some cases form part of the product as a linker group between the antibody of the present disclosure and the polymer.

The size of the polymer can be changed as needed, but the size of the polymer is generally in the range of 500Da to 50000Da, for example, 5000 to 40000Da, for example, an average molecular weight of 20000 to 40000Da. The polymer size can be selected based on the desired use of the product, such as the ability to locate to certain tissues such as tumors or to extend the circulation half-life (for review, see Chapman, 2002, Advanced Drug Delivery Reviews, 54, 531-545). Thus, for example, when it is intended that the product leave the circulation and infiltrate the tissue, for example, for treating tumors, a small molecular weight (e.g., a molecular weight of about 5000Da) polymer can be advantageously used. For applications in which the product remains in the circulation, a polymer with a higher molecular weight can be advantageously used, for example, with a molecular weight in the range of 20000Da to 40000Da.

Suitable polymers include polyalkylene polymers, such as poly(ethylene glycol) or, in particular, methoxypoly(ethylene glycol) or derivatives thereof, in particular having a molecular weight in the range of about 15,000 Da to about 40,000 Da.

In one example, the antibody for use in the present invention is connected to a poly (ethylene glycol) (PEG) moiety. In a specific example, the PEG molecule can be connected by any available amino acid side chain or terminal amino acid functional group (e.g., any free amino group, imino group, sulfhydryl group, hydroxyl group, or carboxyl group) located in the antibody. Such amino acids may be naturally present in antibodies or may be engineered into antibodies using recombinant DNA methods (see, e.g., US 5,219,996; US 5,667,425; WO 98/25971). In one example, the molecule of the present invention is a modified antibody in which the modification is the addition of one or more amino acids to the C-terminus of its heavy chain to allow effector molecules to attach. Multiple sites can be used to connect two or more PEG molecules.

In one embodiment, the PEG molecule is attached to cysteine 171 in the light chain, see, eg, WO 2008/038024 (incorporated herein by reference).

The PEG molecule is covalently linked via the sulfhydryl group of at least one cysteine residue located in the antibody. Each polymer molecule connected to the modified antibody can be covalently linked to the sulfur atom of the cysteine residue located in the antibody. The covalent linkage is typically a disulfide bond or particularly a sulfur-carbon bond. When the sulfhydryl group is used as the connection point for the appropriately activated effector molecule, for example, sulfhydryl selective derivatives such as maleimide and cysteine derivatives can be used. The activated polymer can be used as a starting material in the preparation of the above-mentioned polymer-modified antibodies. The activated polymer can be any polymer containing a sulfhydryl reactive group such as an α-halocarboxylic acid or ester such as iodoacetamide, an imide, such as maleimide, vinyl sulfone or a disulfide. Such starting materials are commercially available (e.g., from Nektar, formerly known as Shearwater Polymers Inc., Huntsville, AL, USA) or can be prepared from commercially available starting materials using conventional chemical methods. Specific PEG molecules include 20K methoxy-PEG-amine (available from Nektar, formerly Shearwater; Rapp Polymere; and SunBio) and M-PEG-SPA (available from Nektar, formerly Shearwater).

The present invention also provides isolated DNA encoding the antibody molecules described herein.

In other aspects, a vector comprising the DNA is provided.

The general methods by which vectors can be constructed, transfection methods and culture methods are well known to those skilled in the art. In this regard, reference is made to "Current Protocols in Molecular Biology", 1999, F. M. Ausubel (ed), Wiley Interscience, New York and the Maniatis Manual (produced by Cold Spring Harbor Publishing).

In other aspects, a host cell comprising the vector and/or DNA is provided.

Any suitable host cell/vector system can be used to express the DNA sequence encoding the molecules of the present invention. Bacteria, such as E. coli and other microbial systems, or eukaryotic (e.g., mammalian) host cell expression systems can also be used. Suitable mammalian host cells include CHO, myeloma, or hybridoma cells.

The invention also provides methods for producing the antibody molecules described herein, comprising culturing a host cell comprising a vector (and/or DNA) of the invention under conditions suitable for expression of protein from DNA encoding the antibody molecule of the invention, and isolating the antibody molecule.

In order to produce a product comprising a heavy chain and a light chain, a cell line can be transfected with two vectors, a first vector encoding a light chain polypeptide and a second vector encoding a heavy chain polypeptide. Alternatively, a single vector can be used that includes sequences encoding both heavy and light chain polypeptides.

Antibody molecules according to the present disclosure are expressed from host cells at appropriate levels, making them amenable to commercial processing.

Antibodies can be specific for any target antigen. Antigens can be cell-bound proteins, such as cell surface proteins on cells such as bacterial cells, yeast cells, T cells, endothelial cells, or tumor cells, or they can be soluble proteins. Target antigens can also be any medically relevant proteins, such as those that are upregulated during disease or infection, such as receptors and/or their corresponding ligands. Specific examples of cell surface proteins include adhesion molecules, such as integrins, such as β1 integrins, such as VLA-4, E-selectin, P-selectin, or L-selectin, CD2, CD3, CD4, CD5, CD7, CD8, CD11a, CD11b, CD18, CD19, CD20, CD23, CD25, CD33, CD38, CD40, CD40L, CD45, CDW52, CD69, CD134 (OX40), ICOS, BCMP7, CD137, CD27L, CDCP1, CSF1 or CSF1-receptor, DPCR1, DPCR1, dudulin2, FLJ20584, FLJ40787, HEK2, KIAA0634, KIAA0659, KIAA1246, KIAA1455, LTBP2, LTK, MAL2, MRP2, integrin-like 2, NKCC1, PTK7, RAIG1, TCAM1, SC6, BCMP101, BCMP84, BCMP11, DTD, carcinoembryonic antigen (CEA), human milk fat globulins (HMFG1 and 2), MHC class I and MHC class II antigens, KDR and VEGF, PD-1, DC-SIGN, TL1A, DR3, IL-7 receptor A and, where appropriate, their receptors.

Soluble antigens include interleukins such as IL-1, IL-2, IL-3, IL-4, IL-5, IL-6, IL-8, IL-12, IL-13, IL-14, IL-16 or IL-17, such as IL17A and/or IL17F, viral antigens such as respiratory syncytial virus or cytomegalovirus antigens, immunoglobulins such as IgE, interferons such as interferon α, interferon β or interferon γ, tumor necrosis factor TNF (formerly known as tumor necrosis factor-α, herein referred to as TNF or TNFα), tumor necrosis factor-β, colony stimulating factors such as G-CSF or GM-CSF, and platelet-derived growth factors such as PDGF-α and PDGF-β, WISP-1 and, where appropriate, their receptors. Other antigens include bacterial cell surface antigens, bacterial toxins, viruses such as influenza virus, EBV, HepA, B and C, bioterrorism agents, radionuclides and heavy metals, and snake and spider venom and toxins.

In one embodiment, the antibody can be used to functionally alter the activity of the target antigen. For example, the antibody can directly or indirectly neutralize, antagonize, or agonize the activity of the antigen.

In one embodiment, the disclosure extends to a method for producing an asymmetric hybrid antibody according to the disclosure, the method comprising the steps of obtaining a symmetric antibody (i.e., an antibody wherein the two heavy chains are identical/identical) comprising a first heavy chain sequence or a fragment thereof as defined herein, mixing the antibody in vitro with a second symmetric antibody comprising a second heavy chain sequence or a fragment thereof, wherein the second heavy chain sequence is different from the first heavy chain sequence, under conditions that favor heavy chain exchange between the two antibodies, and optionally isolating the asymmetric hybrid antibody.

In vitro conditions that facilitate dynamic exchange include reducing conditions. Suitable reducing agents include GSH, 2-mercaptoethanol, 2-mercaptoethylamine, TBP, TCEP, cysteine-HCl, and DTT.

Suitable concentrations of reducing agents are in the range of 0.01 to 10 mM, for example 0.5 to 5 mM. Furthermore, redox buffers may be used to achieve reduction, ie different relative ratios of the oxidizing and reducing versions of the reagents, for example: GSH:GSSG and Cys:diCys.

Suitable conditions include a ratio of antibodies of 0.5:5 to 5:05, for example 1:1 or 1:2.

Suitable temperatures include 15 to 40°C, for example 37°C.

Reducing conditions can be selected that lie between the reduction stabilities of the homodimer and the heterodimer.

In an alternative embodiment, if the antibodies of the present disclosure are produced using mixed cell cultures, there is, for example, about a 50% exchange. This can yield approximately 1-2 g/l of the desired bispecific antibody.

In one embodiment, asymmetric antibodies obtained or obtainable by the methods described herein and formulations comprising the same are provided, particularly for use in therapy.

The antibody molecules of the present invention are useful in the treatment and/or prevention of pathological conditions.

Thus, there is provided an antibody according to the invention for use in therapy by administering a therapeutically effective amount thereof, eg in a pharmaceutical formulation. In one embodiment, the antibody according to the invention is administered topically to the lung, eg by inhalation.

The antibodies provided herein are useful in the treatment of diseases or disorders, including inflammatory diseases and disorders, immune diseases and disorders, fibrotic disorders, and cancer.

The terms "inflammatory disease" or "disorder" and "immune disease or disorder" include rheumatoid arthritis, psoriatic arthritis, Still's disease, Muckle-Wells disease, psoriasis, Crohn's disease, ulcerative colitis, SLE (systemic lupus erythematosus), asthma, allergic rhinitis, atopic dermatitis, multiple sclerosis, vasculitis, type I diabetes, transplantation, and graft-versus-host disease.

The term "fibrotic disorder" includes idiopathic pulmonary fibrosis (IPF), systemic sclerosis (or scleroderma), renal fibrosis, diabetic nephropathy, IgA nephropathy, hypertension, end-stage renal disease, peritoneal fibrosis (continuous ambulatory peritoneal dialysis), cirrhosis, age-related macular degeneration (ARMD), retinopathy, cardiac reactive fibrosis, scars, keloids, burns, skin ulcers, angioplasty, coronary artery bypass graft surgery, arthroplasty, and cataract surgery.

The term "cancer" includes malignant neoplasms arising from the epithelium found in the skin or, more commonly, the lining of body organs such as the breast, ovary, prostate, lung, kidney, pancreas, stomach, bladder, or intestine. Cancer tends to infiltrate adjacent tissues and spread (metastasize) to distant organs, for example, to the bones, liver, lungs, or brain.

The present invention also provides pharmaceutical or diagnostic compositions comprising an antibody of the present invention in combination with one or more pharmaceutically acceptable excipients, diluents, or carriers. Thus, use of the antibodies of the present invention for the manufacture of a medicament is provided. The compositions are typically provided as part of a sterile pharmaceutical composition, which typically includes a pharmaceutically acceptable carrier. The pharmaceutical compositions of the present invention may additionally include a pharmaceutically acceptable adjuvant.

The present invention also provides a method for preparing a pharmaceutical or diagnostic composition, comprising adding and mixing an antibody of the present invention with one or more pharmaceutically acceptable excipients, diluents or carriers.

The antibodies of the present disclosure may be the sole active ingredient in a pharmaceutical or diagnostic composition or may be accompanied by other active ingredients, including other antibody components, such as anti-TNF, anti-IL-1β, anti-T cell, anti-IFNγ or anti-LPS antibodies or non-antibody components such as xanthines. Other suitable active ingredients include antibodies capable of inducing tolerance, such as anti-CD3 or anti-CD4 antibodies.

In other embodiments, antibodies or compositions according to the present disclosure are used in combination with other pharmaceutically active agents such as corticosteroids (e.g., fluticasone propionate) and/or beta-2-agonists (e.g., albuterol, salmeterol, or formoterol) or inhibitors of cell growth and proliferation (e.g., rapamycin, cyclophosphamide, methotrexate), or alternatively, CD28 and/or CD40 inhibitors. In one embodiment, the inhibitor is a small molecule. In another embodiment, the inhibitor is an antibody specific for the target.

The pharmaceutical composition suitably comprises a therapeutically effective amount of an antibody of the present invention. As used herein, the term "therapeutically effective amount" refers to the amount of therapeutic agent required to treat, alleviate or prevent a targeted disease or condition, or to show a detectable therapeutic or preventive effect. A therapeutically effective amount can be initially evaluated in cell culture assays or in animal models, typically in rodents, rabbits, dogs, pigs or primates. Animal models can also be used to determine appropriate concentration ranges and routes of administration. Such information can then be used to determine useful doses and routes for administration in humans.

The precise therapeutically effective amount for a human subject will depend on the severity of the disease state, the subject's general health, the subject's age, weight and sex, diet, time and frequency of administration, drug combination, reaction sensitivity and tolerance/response to treatment. Such an amount can be determined by routine experimentation and is within the discretion of the clinician. Typically, a therapeutically effective amount is from 0.01 mg/kg to 50 mg/kg, for example, from 0.1 mg/kg to 20 mg/kg. The pharmaceutical composition can be conveniently provided in a unit dosage form containing a predetermined amount of the active agent of the present invention per dose.

The composition can be administered to the patient alone or can be administered in combination (eg, simultaneously, sequentially, or separately) with other agents, drugs, or hormones.

The dosage of an antibody of the invention administered will depend on the nature of the condition to be treated, eg the extent of disease/inflammation present and on whether the molecule is being used prophylactically or to treat an existing condition.

The frequency of administration will depend on the half-life of the antibody and its duration of action. If the antibody has a short half-life (e.g., 2 to 10 hours), it may be necessary to provide one or more doses daily. Alternatively, if the antibody has a long half-life (e.g., 2 to 15 days), it may only be necessary to provide 1 dose once a day, once a week, or once every 1 or 2 months.

The pharmaceutically acceptable carrier should not itself induce the production of antibodies harmful to the individual receiving the composition and should not be toxic. Suitable carriers can be large, slowly metabolized macromolecules such as proteins, polypeptides, liposomes, polysaccharides, polylactic acids, polyglycolic acids, polyamino acids, amino acid copolymers and inactivated virus particles.

Pharmaceutically acceptable salts may be used, such as mineral acid salts, for example hydrochlorides, hydrobromides, phosphates and sulfates, or salts of organic acids, for example acetates, propionates, malonates and benzoates.

Pharmaceutically acceptable carriers in therapeutic compositions may additionally comprise liquids such as water, saline, glycerol, and ethanol. In addition, auxiliary substances, such as wetting agents or emulsifiers or pH buffering substances may be present in such compositions. Such carriers enable pharmaceutical compositions to be formulated as tablets, pills, lozenges, capsules, liquids, gels, syrups, ointments, and suspensions for ingestion by patients.

Suitable administration forms include forms suitable for parenteral administration, such as by injection or infusion, for example by single rapid intravenous injection or continuous infusion. When the product is intended for injection or infusion, it may take the form of a suspension, solution or emulsion in an oily or aqueous vehicle, which may contain preparatants such as suspending agents, preservatives, stabilizers and/or dispersants. Alternatively, the molecules of the present disclosure may be present in a dry form for reconstitution with an appropriate sterile liquid prior to use.

After formulation, the compositions of the present invention can be administered directly to a subject. The subject to be treated can be an animal. However, in one or more embodiments, the compositions are suitable for administration to a human subject.

Suitably in formulations according to the present disclosure, the pH of the final formulation is not similar to the value of the isoelectric point of the antibody, for example if the pH of the formulation is 7, a pi of 8-9 or higher may be appropriate. While not wishing to be bound by theory, it is thought that this may ultimately provide a final formulation with enhanced stability, for example the antibody remains in solution.

The pharmaceutical composition of the present invention can be used by any number of approaches, including but not limited to oral, intravenous, intramuscular, intraarterial, intramedullary, intradural, intraventricular, percutaneous, percutaneous (e.g., referring to WO98/20734), subcutaneous, intraperitoneal, intranasal, intestinal, local, sublingual, intravaginal or rectal approaches. Hypodermic syringes can also be used for administering the pharmaceutical composition of the present invention. Typically, therapeutic compositions can be prepared as injectables, such as liquid solutions or suspensions. Solid forms suitable for dissolving or being suspended in a liquid vehicle before injection can also be prepared.

Direct delivery of the composition is typically achieved by subcutaneous, intraperitoneal, intravenous or intramuscular injection, or delivered to the interstitial space of the tissue. The composition can also be applied to a wound. Dosage therapy can be a single dose regimen or a multiple dose regimen.

It will be understood that the active ingredient in the composition will be an antibody. As such, it will be susceptible to degradation in the gastrointestinal tract. Therefore, if the composition is to be administered via a route using the gastrointestinal tract, the composition will need to contain an agent that protects the antibody from degradation but releases the antibody once it is absorbed by the gastrointestinal tract.

A thorough discussion of pharmaceutically acceptable carriers can be found in Remington's Pharmaceutical Sciences (Mack Publishing Company, NJ 1991).

In one embodiment, the formulation is provided as a formulation for topical administration, including inhalation.

Suitable inhalable formulations include inhalable powders, metered dose aerosols containing a propellant gas or inhalable solutions without a propellant gas. The inhalable powders according to the present disclosure containing the active substance may consist solely of the active substance described above or of a mixture of the active substance described above with a physiologically acceptable excipient.

These inhalable powders may comprise monosaccharides (e.g. glucose or arabinose), disaccharides (e.g. lactose, sucrose, maltose), oligosaccharides and polysaccharides (e.g. dextran), polyols (e.g. sorbitol, mannitol, xylitol), salts (e.g. sodium chloride, calcium carbonate) or mixtures thereof. Monosaccharides or disaccharides may be suitably used, lactose or glucose being used, particularly but not exclusively in the form of their hydrates.

Particles for deposition in the lungs require a particle size of less than 10 microns, such as 1-9 microns, for example 0.1 to 5 μm, especially 1 to 5 μm. The particle size of the active ingredient (eg antibody) is of primary importance.

Propellant gases that can be used to prepare inhalable aerosols are known in the art. Suitable propellant gases are selected from hydrocarbons such as n-propane, n-butane or isobutane, and halogenated hydrocarbons such as chlorinated and/or fluorinated derivatives of methane, ethane, propane, butane, cyclopropane or cyclobutane. The above-mentioned propellant gases can be used alone or in mixtures thereof.

Particularly suitable propellant gases are halogenated hydrocarbon derivatives selected from TG11, TG12, TG134a and TG227. Of the above halogenated hydrocarbons, TG134a (1,1,1,2-tetrafluoroethane) and TG227 (1,1,1,2,3,3,3-heptafluoropropane) and mixtures thereof are particularly suitable.

The propellant-containing inhalable aerosol may also contain other ingredients such as co-solvents, stabilizers, surfactants (surface active agents), antioxidants, lubricants and means for adjusting pH. All these ingredients are known in the art.

The propellant-gas-containing inhalable aerosol according to the invention may contain up to 5% by weight of active substance. The aerosol according to the invention contains, for example, 0.002 to 5% by weight, 0.01 to 3% by weight, 0.015 to 2% by weight, 0.1 to 2% by weight, 0.5 to 2% by weight or 0.5 to 1% by weight of active ingredient.

Alternatively, local administration to the lungs can also be by administration of a liquid solution or suspension formulation, for example, using a device such as a nebulizer, for example, a nebulizer connected to a compressor (e.g., the Pari LC-Jet Plus(R) nebulizer connected to a Pari Master(R) compressor manufactured by Pari Respiratory Equipment, Inc., Richmond, Va.).

The antibodies of the present invention can be delivered dispersed in a solvent, for example, in the form of a solution or suspension. They can be suspended in an appropriate physiological solution, such as saline or other pharmaceutically acceptable solvent or buffer. Buffer solutions known in the art can contain 0.05 mg to 0.15 mg of disodium edetate, 8.0 mg to 9.0 mg of NaCl, 0.15 mg to 0.25 mg of polysorbate, 0.25 mg to 0.30 mg of anhydrous citric acid, and 0.45 mg to 0.55 mg of sodium citrate per 1 mL of water to achieve a pH of about 4.0 to 5.0. Suspensions can use, for example, lyophilized molecules.

Therapeutic suspension or solution formulations may also include one or more excipients. Excipients are well known in the art and include buffers (e.g., citrate buffer, phosphate buffer, acetate buffer, and bicarbonate buffer), amino acids, urea, alcohol, ascorbic acid, phospholipids, proteins (e.g., serum albumin), EDTA, sodium chloride, liposomes, mannitol, sorbitol, and glycerol. The solution or suspension may be encapsulated in liposomes or biodegradable microspheres. Preparations are typically provided in a substantially sterile form (using an aseptic manufacturing process).

This may include production and sterilization by filtering the buffer solvent/solution used for the formulation, sterile suspension of the molecule in a sterile buffer solvent solution, and dispensing the formulation into sterile containers using methods well known to those skilled in the art.

The sprayable formulations according to the present disclosure can be provided, for example, in single dosage units (e.g., sealed plastic containers or vials) packaged in foil envelopes. Each vial contains a unit dose in a volume (e.g., 2 ml) of solvent/solution buffer.

Antibodies according to the present disclosure are believed to be particularly suitable for delivery by spray.

In the context of this specification, comprising means including.

Technically suitable embodiments of the present invention can be combined therein.

"Asymmetry" and "asymmetric hybrid" are used interchangeably herein.

The present invention will be described with reference to the following examples, which are merely illustrative and should not be construed in any way as limiting the scope of the invention.

Example

1. Mutagenesis of IgG4 Heavy Chain and Generation of Single-Gene Vectors of Mutated IgG4 Heavy Chain

Amino acid mutations were performed using the Lightening Multi Site Directed Mutagenesis (SDM) kit or the BIOMEDIA® II DSM kit (available from BIOMEDIA®) (Cat. Nos. 210516 and 200521, respectively) according to the manufacturer's instructions.

The mutations were verified by DNA sequencing. The IgG4 heavy chains of antibodies 1 to 47 in the following table were generated:

Other antibodies prepared are described in the table above.

The heavy chain of antibody 48 (SEQ ID NO: 266) was generated by PCR and restriction enzyme cloning. The PCR product was generated using a forward oligonucleotide encoding the IgG1 upper hinge and core hinge region sequences and the restriction site BglII and a reverse oligonucleotide encoding the restriction enzyme DraIII. The PCR fragment was then digested with the above enzymes and ligated into the hG4 single gene vector containing the appropriate variable region.

2. Expression of Mutated IgG4 Antibodies

All mutant DNAs were transfected into CHOK1 cells. Cells (2x10 ⁸ cells/ml) were resuspended in 1 ml of Earle's balanced salt solution (Sigma) and mixed with 400 μg of DNA (200 μg heavy chain DNA and 200 μg kappa light chain DNA). 800 μl aliquots were transferred to 0.4 cm cuvettes (Biorad). For a 500 ml culture, 6 cuvettes were electroporated under the following parameters: 1 ms, 9.6 Amps; 10 ms, 0 Amps; 40 ms, 3.2 Amps. The transfected cells were incubated at 37 ° C for 24 hours in a 5% CO ₂ humidified environment with shaking at 140 rpm, and incubated at 32 ° C for 10-13 days from the second day after transfection. On the fourth day after transfection, 1.6 ml of 1 M sodium butyrate was added to the culture. When the cells reached a viability of 40% or on the 13th day, the supernatant was collected. The culture was centrifuged at 4000 rpm for 45 minutes and the supernatant was passed through a 0.22 μM Stericup filter (Millipore) for purification.

3. Purification of Mutated IgG4 Antibodies

The supernatant (200-500 ml) was purified using a Protein A 5 ml HiTrap MabSelect SuRe column (GE Healthcare, Amersham UK). Samples were prepared by adding 1/50 of the supernatant volume of 2M Tris-HCl pH 8.5. The sample was loaded onto the column at a rate of 1 ml/min. The column was washed with PBS pH 7.4. To elute the sample, 0.1 M sodium citrate pH 3.4 was passed through the column at 1 ml/min, collecting 0.5 ml fractions. Peak fractions were neutralized by adding 0.125 ml of 2M Tris-HCl pH 8.5 to each fraction. UV detection was set at 280 nm.

4. Characterization of Purified Mutated IgG4 Antibodies

SDS PAGE analysis:

The crude supernatant was centrifuged at 1200 rpm for 5 min and subsequently quantified on an OCTET. Antibody samples (25-30 ng) were prepared by adding an appropriate amount of antibody, 4x loading buffer (Invitrogen), and 2 μl of 100 mM NEM. A total volume of 20 μl was prepared using dH ₂ O. The sample was then boiled at 100° C. for 3 min and loaded onto a 1.5 mm 4-20% Tris-glycine gel in 15 wells. The gel was electrophoresed at 150 V in 1x Tank buffer for 1.5 hours. The antibody was transferred to a nitrocellulose membrane using an iBlot dry transfer system device, and the transfer was performed for 8 min. The membrane was incubated in PBS-TM at room temperature (RT) for 1 hour on a shaking platform, and then incubated with rabbit anti-human IgG Fc HRP-conjugated antibody (Jackson Immunoresearch) or goat anti-human κ light chain HRP-conjugated antibody (Bethyl) at RT for 1 hour while shaking. This was followed by three 5-minute washes with PBS-T. The blots were developed using the Metal Enhanced DAB Substrate Kit according to the manufacturer's instructions (Pierce).

The results of immunoblotting analysis are shown in Figures 7, 8, 9 and 10. In Figures 7-10, H chain represents heavy chain, L represents light chain, H2L2 is a complete antibody molecule containing two heavy chains and two light chains, and HL is a half molecule containing one heavy chain and one light chain.

Figure 7 shows immunoblot analysis of antibodies 15, 16, 6, 7, 8, 17, 18, 19, 5, 5P, 9, 10, 11, 1, 2, 3, 4, 12, 13, and 14. As can be seen in Figure 7, the antibodies exhibited good levels of H2L2, with the exception of antibodies 4, 8, and 14, which exhibited little or no H2L2 due to the presence of two hinge mutations, C239S and C242S. However, antibodies 4, 8, and 14 were able to form H2L2 through non-covalent bonding between heavy chains. Mutant 3 also exhibited minimal H2L2, which retained C239 but was unable to form intra-heavy chain disulfides in the hinge. This is presumably due to efficient disulfide formation between the C-terminal light chain (LC) cysteine and hinge C239. It can also be seen that antibodies containing the mutation C239S but not C242S (antibodies 2, 6, 9, and 12) show reduced HL formation compared to antibodies containing neither C239S nor C242S or antibodies containing C242S but not C239S. Antibodies 5P and 16 containing the S241P mutation also show reduced HL formation. Comparison of mutants 2 and 3 shows the extent of "reach" of the C-terminal cysteine of the light chain that forms a disulfide bond with the heavy chain, with the light chain cysteine appearing to bond more efficiently to C239 in the heavy chain than to C242.

Figure 8 shows immunoblot analysis of antibodies 15, 6, 7, 8, 28, 29, 30, 31, 17, 19, 32, 33, 33, 34, 35, 36, 37, 38, and 39. As can be seen from Figure 8, with the exception of antibodies 8, 31, 35, and 39 (which exhibit no or minimal H2L2 due to the presence of mutations C239S and C242S in the hinge region and thus no disulfide bond formation between the two heavy chains), the antibodies exhibit good levels of H2L2. However, antibodies 8, 31, 35, and 39 can form H2L2 through non-covalent bonding between the heavy chains. It can also be seen that antibodies comprising mutation C239S but not comprising C242S (antibodies 6, 29, 33 and 37) show the formation of reduced HL compared to antibodies neither comprising C239S nor comprising C242S or antibodies comprising C242S but not comprising C239S. Mutant 15 is able to form disulfide bonds and inter-heavy chain disulfide between the light chain and the G230C in CH1, thereby leading to a fully assembled and disulfide-bonded protein. In addition, the position of the cysteine introduced in the upper hinge of the comparison display of mutants 15 (C127S G230C), 28 (C127S Y229C), 32 (C127S K228C) and 36 (C127S S227C) increases the disulfide bond formation between LC-HC. G230 and Y229 are particularly preferred positions for introducing cysteine. Mutant 28 (C127S Y229C) showed low levels of HL and H2, thus having low disulfide heterogeneity.

Figure 9 shows immunoblot analysis of antibodies 15, 6, 7, 8, 44, 45, 46, 47, 17, and 19. As can be seen from Figure 9, with the exception of antibodies 8 and 47 (which exhibit no or minimal H2L2 due to the presence of C239S and C242S mutations in the hinge region and the resulting lack of disulfide bond formation between the two heavy chains), the antibodies exhibit good levels of H2L2. However, antibodies 8 and 47 can form H2L2 through non-covalent bonding between the heavy chains. It can also be seen that antibodies containing the mutation C239S but not C242S (antibodies 6 and 45) exhibit reduced HL formation compared to antibodies not containing C239S and C242S or antibodies containing C242S but not C239S. In particular, mutation 44 showed that the insertion of 3 amino acids in the upper hinge could also reduce the formation of HL and H2, resulting in a lower level of disulfide heterogeneity than the comparable mutant 15.

Figure 10 shows immunoblot analysis of antibodies 48, 17, 18, and 19. As can be seen in Figure 10, antibody 48 exhibits good levels of H2L2 and minimal HL. Mutant 48 comprises the IgG1 upper hinge sequence EPKSCDKTHT SEQ ID NO: 319 (replacing the IgG4 upper hinge sequence) and a core hinge S241P mutation. Thus, mutant 48 has both the upper and core hinge sequences EPKSCDKTHTCPPCP SEQ ID NO: 320. Mutant 48 exhibits lower levels of disulfide heterogeneity compared to wild-type IgG4 antibody 17 and approximately equal levels of disulfide heterogeneity compared to IgG4 S241P mutant 18 and wild-type IgG1 antibody 19.

Thermofluor assay:

The thermal stability of the purified mAb was analyzed in a thermofluor assay by monitoring the thermal unfolding of the protein using Orange. 5 μl of 1 mg/ml mAb, 5 μl of 30x dye, and 40 μl of PBS were added together. 10 μl of the mixture was distributed in quadruplicate to a 384 PCR optical well plate, and PCR was subsequently performed on a 7900HT Fast Real-Time PCR system (Agilent Technologies UK Ltd, Wokingham UK). The PCR system includes a heating device for precise temperature control set at 20°C to 99°C; a charge-coupled device was used to monitor fluorescence changes in the wells.

Figures 11, 12, 13, 14 and 15 show the results of thermal stability analysis of IgG4 antibody mutants compared to wild-type IgG1 and IgG4 antibodies.

The comparison of mutant 15 and wild-type IgG4 (mutant 17) shows an increase in the Fab Tm caused by the disulfide arrangement of the change. The comparison of mutant 15 and 28 shows a further increase in the Fab Tm of mutant 28 comprising the Y229C mutation compared to the mutant 15 comprising the G230C mutation. The comparison of mutant 15 and 44 shows that the Fab Tm of the G230C mutant can be further increased by inserting 3 amino acids in the upper hinge. The comparison of mutant 17 and 18 shows that the S241P hinge mutation does not increase Fab Tm, even if it significantly reduces HL formation. When compared with wild-type IgG4 (mutant 17) and mutant 15, mutant 48 also shows a significantly increased Fab Tm.

Figure 15 shows the overall ranking of the thermal stability of selected IgG4 mutants according to the present invention. Mutants 48, 44, 44P, 46, 45, 6, 29, 30, 28, 28P, 31, 8, 47 and 15 all have significantly higher Fab Tm values than wild-type IgG4 (mutant 17) and wild-type IgG4 S241P (mutant 18).

5.Fab arm exchange

a) In vitro heavy chain exchange

A first IgG4 antibody and a second IgG4 antibody, each with a different antigenic specificity, were mixed in a phosphate-buffered saline (PBS) solution (in mM: 150 NaCl, 10 NaH ₂ PO ₄ ; pH 7.4) at a total concentration of 100 ug/ml in a 1:2 molar ratio. To reduce disulfide bonds, samples were supplemented with reduced glutathione (GSH; Sigma) to a final concentration of 0, 0.5, or 5 mM. At the start of the experiment (t = 0 hours), an aliquot of the mixture was removed and quenched with N-ethylmaleimide (NEM; Sigma) to a final concentration of 10 mM (to inactivate potentially reactive sulfhydryl groups), and incubated side by side with the remaining mixture at 37° C. for 16 hours (t = 16 hours). Following incubation, the sample at t = 16 hours was quenched as described above.

The combinations of primary and secondary antibodies tested are shown in the table below:

Heavy chain swapping between Antibodies 1 and 2 in the table above provides an asymmetric antibody having heavy chains from each of the related antibodies.

b) Detection and quantification of heavy chain exchange in vitro

The presence of functionally active bispecific antibodies was determined using a sandwich MSD assay. In this assay, the quenched reaction sample provided in Example 5a) was pre-incubated with 1 μg/ml of biotinylated antigen 1 (the antigen of the primary antibody) in PB for 1 hour at room temperature with serial dilutions of 1% BSA (PB) in PBS, under stirring (200 r.p.m.) and then transferred to a streptavidin-coated MSD plate (Meso Scale Diagnostics) pre-blocked with PB. After incubation for 1 hour at room temperature with stirring, the wells were washed three times with PBS/0.1% Tween-20 and then incubated with 1 μg/ml of thio-labeled antigen 2 (the antigen of the secondary antibody) in PB. After incubation, the plates were washed as described above, and the signals were visualized and measured using the manufacturer's read buffer and an Image Sector 6000 instrument, respectively. Background values obtained from control replicates in which the biotinylated antigen was replaced with a non-biotinylated surrogate were subtracted from all signals. Duplicate values from at least 3 independent experiments were used in all calculations.The higher the MSD signal, the greater the amount of heavy chain exchange that has occurred.

Figure 16 shows heavy chain exchange at 16 hours under two GSH concentrations, wherein the first antibody is selected from IgG1 wild-type, IgG4 wild-type and various mutant antibodies and the second antibody is IgG4 wild-type. The figure shows that the mutants have less exchange than the wild-type IgG4 antibody and significantly more exchange than the IgG1 wild-type antibody. This is advantageous because exchange can be used to prepare asymmetric antibodies of the present disclosure in vitro, which have a lower susceptibility to exchange in vivo than the wild-type IgG4 antibody. In some cases, increasing the concentration of a reducing agent such as GSH increases the amount of exchange observed.

In good agreement with the literature (Labrijn 2011, Lewis 2009, Stubenrauch 2010, Labrijn 2009), we show that the S241P mutation in the IgG4 core hinge represents a tool for preventing Fab-arm exchange (Figure 16). It can also be seen that the mutant bispecific antibodies of the present invention show less Fab arm exchange than at 0.5mM GSH (a concentration that is 100 times higher than the physiological GSH concentration of 4-6uM plasma (Zilmer et al., 2005. Drug Design Reviews). Thus, bispecific antibodies can be generated in vitro by Fab arm exchange under reducing conditions, which then have significantly reduced in vivo Fab arm exchange compared to IgG4wt.

Figure 17 shows that the glycine at position 241 can be easily exchanged with IgG4 mutants having an alanine or threonine at this position. The exchange of IgG4 with alanine at position 241 and a mutant with threonine at this position will be more than the exchange with a mutant with glycine at this position. Similarly, in the reaction with S241G, the IgG4 mutant with threonine at position 241 showed reduced exchange activity compared to the symmetry assay. The exchange with IgG4S241A was similar to the symmetry assay. In short, this shows that IgG4S241T exchanges at a level similar to IgG4WT and is more likely to exchange than mutants S241A and S241G.

Antibody Affinity:

The affinity of selected mutant IgG4 antibodies of the present invention for soluble target cytokines can be measured using BIAiacore ^™ . The assay format is capture of IgG on anti-Fc expression followed by titration of soluble cytokines.

The term " _kd " (s ^-1 ) refers to the dissociation rate constant of an antibody-antigen interaction. This value is also referred to as the _koff value.

As used herein, the term " _ka " (M ^"1s " ¹ ) refers to the association rate constant for an antibody-antigen interaction.

As used herein, the term " _KD " (M) or " _KD " (pM) refers to the dissociation equilibrium constant for an antibody-antigen interaction.

Size Exclusion (SEC) HPLC Analysis:

Approximately 50 μg of purified antibody was run on an HPLC column using an S200 column. Abs 1 to 19 were used for analysis. The results showed that non-covalently associated H2L2 was formed, although the DSB arrangement of the human IgG4 molecule was altered.

Claims (36)

1. An asymmetric antibody comprising two heavy chains or heavy chain fragments, each comprising at least a variable region, a hinge region, and a _CH1 domain, wherein the first heavy chain or fragment thereof is of class IgG4 and has: In the _aCH1 domain, the interchain cysteine residue at position 127 (Kabat numbering system) is replaced by another amino acid; and b. One or more amino acids located in the upper hinge region at positions 227, 228, 229, and 230 (numbered according to the Kabat numbering system) are replaced by cysteine, and The second heavy chain or a fragment thereof is characterized in that part or all of the chain has an amino acid sequence different from that of the first heavy chain in at least a constant region, and wherein the constant sequence of the second heavy chain is IgG4 wild type or IgG4 wild type with S241P mutation. 2. An asymmetric antibody comprising a first and a second heavy chain or heavy chain fragment, each comprising at least a variable region, a hinge region, and a _CH1 domain, wherein the first heavy chain or fragment is IgG4 and has an IgG1-type hinge, and the second heavy chain or fragment has an amino acid sequence different from that of the first heavy chain in at least a constant region, wherein the constant sequence of the second heavy chain is IgG4 wild-type or IgG4 wild-type with an S241P mutation, wherein one or more amino acids in the upper hinge region of the first heavy chain are replaced by cysteine residues and selected from amino acids at positions 227, 228, 229, and 230 according to the Kabat numbering system. 3. The asymmetric antibody of claim 1 or 2, wherein the glycine at position 230, numbered according to the Kabat numbering system, is replaced by cysteine. 4. The asymmetric antibody of claim 1 or 2, wherein the tyrosine at position 229, numbered according to the Kabat numbering system, is replaced by a cysteine. 5. The asymmetric antibody of claim 1 or 2, wherein the lysine at position 228, numbered according to the Kabat numbering system, is replaced by a cysteine. 6. The asymmetric antibody of claim 1 or 2, wherein the serine at position 227 according to the Kabat numbering system is replaced by a cysteine. 7. The asymmetric antibody of claim 1 or 2, wherein the cysteine residues at position 239 and position 242 of the first heavy chain, numbered according to the Kabat numbering system, are replaced by another amino acid. 8. The asymmetric antibody of claim 1 or 2, wherein the cysteine at position 239 of the first heavy chain, numbered according to the Kabat numbering system, is replaced by another amino acid. 9. The asymmetric antibody of claim 1 or 2, wherein the cysteine at position 242 in the first heavy chain, numbered according to the Kabat numbering system, is replaced by another amino acid. 10. An asymmetric antibody comprising two heavy chains, each comprising at least a variable region, a hinge region, and a _CH1 domain, wherein the first heavy chain or a fragment thereof is characterized as being IgG4 and having: a) According to the Kabat numbering system, the interchain cysteine at position 127 is replaced by another amino acid; and b) The cysteine at position 239 or position 242, according to the Kabat numbering system, is replaced by another amino acid; wherein the second heavy chain or a fragment thereof is characterized in that it has an amino acid sequence different from that of the first heavy chain in the constant region, and wherein the constant sequence of the second heavy chain is IgG4 wild-type or IgG4 wild-type with the S241P mutation. 11. The asymmetric antibody of claim 10, wherein the cysteine residue at position 239 and/or position 242 of the first heavy chain is replaced by an amino acid that does not contain a thiol group. 12. The asymmetric antibody of claim 11, wherein the amino acid that does not contain a thiol group is serine. 13. The asymmetric antibody of any one of claims 1, 2 or 10-12, wherein the amino acid on residues 240 and/or 241 numbered according to the Kabat numbering system is replaced by another amino acid. 14. The asymmetric antibody of any one of claims 1, 2 or 10-12, wherein the cysteine at position 127 is replaced by an amino acid that does not contain a thiol group. 15. The asymmetric antibody of claim 14, wherein the amino acid that does not contain a thiol group is serine. 16. The asymmetric antibody of any one of claims 1, 2 or 10-12, wherein the heavy chain is mutated to insert 3 amino acids between amino acid 226 and 243 numbered according to the Kabat numbering system. 17. The asymmetric antibody of claim 16, wherein the first heavy chain is mutated to insert three amino acids between positions 238 and 239 according to the Kabat numbering system. 18. The asymmetric antibody of claim 17, wherein three alanine residues are inserted between positions 238 and 239 of the first heavy chain according to the Kabat numbering system. 19. The asymmetric bispecific antibody of claim 17, wherein a threonine, a histidine, and another threonine are inserted between positions 238 and 239 of the first heavy chain according to the Kabat numbering system. 20. The asymmetric antibody of any one of claims 1, 2 or 10-12, wherein the serine at position 241 of the first heavy chain, numbered according to the Kabat numbering system, is replaced by a proline. 21. The asymmetric antibody of any one of claims 1, 2, or 10-12, wherein in the first heavy chain, glycine at position 230 is replaced by cysteine, serine at position 227 is replaced by proline, tyrosine at position 229 is replaced by serine, proline at position 237 is replaced by aspartic acid, proline at position 238 is replaced by lysine, a threonine-histidine-threonine amino acid sequence is inserted between positions 238 and 239, and serine at position 241 is replaced by proline. 22. The asymmetric antibody of any one of claims 1, 2 or 10-12, wherein one or more heavy chains or fragments thereof comprise a _CH2 domain and/or a _CH3 domain. 23. The asymmetric antibody of any one of claims 1, 2 or 10-12, wherein each heavy chain contains a variable region and the two variable regions are identical. 24. The asymmetric antibody of any one of claims 1, 2 or 10-12, wherein each heavy chain contains a variable region and the two variable regions are distinct. 25. The asymmetric antibody of any one of claims 1, 2 or 10-12, wherein the antibody is bispecific. 26. The asymmetric antibody of any one of claims 1, 2 or 10-12, wherein the hinge region in each heavy chain or heavy chain segment is different. 27. The asymmetric antibody of any one of claims 1, 2 or 10 to 12, wherein the hinge region in each heavy chain or heavy chain segment is identical. 28. The asymmetric antibody of any one of claims 1, 2 or 10 to 12, wherein the heavy chain or each heavy chain comprises an upper hinge region and a core region of 12 to 17 amino acids in length. 29. The asymmetric antibody of claim 28, wherein the length of the upper hinge region and the core region is 15 amino acids. 30. The asymmetric antibody of any one of claims 1, 2 or 10-12, wherein the second heavy chain is class IgG4 or class IgG1. 31. The asymmetric antibody of claim 30, wherein the second heavy chain is IgG4. 32. The antibody as defined in any one of claims 1, 2 or 10-12, for the treatment of a disease or disorder. 33. An expression vector comprising a sequence encoding an antibody as defined in any one of claims 1 to 32. 34. A host cell comprising the vector as defined in claim 33. 35. Use of the antibody as defined in any one of claims 1 to 32 in the preparation of a medicament for treating a disease or disorder. 36. A method for producing an antibody as defined in any one of claims 1 to 32, comprising the steps of: obtaining a symmetrical antibody comprising a first heavy chain sequence or a fragment thereof as defined herein; mixing the antibody in vitro with a second symmetrical antibody comprising a second heavy chain sequence or a fragment thereof that differs from the first heavy chain sequence at least in a constant region, under conditions conducive to heavy chain exchange between the two antibodies; and optionally separating the asymmetrical mixed antibody obtained therefrom.