IgA Fc-IgG Fc tandem protein construct

Abstract

The present invention provides a protein construct comprising a human IgA Fc region and a human IgG Fc region, wherein the C-terminus of the human IgA Fc region is linked to the N-terminus of the human IgG Fc region. Such a construct may further comprise a targeting domain. Compositions comprising the construct and therapeutic uses of the construct are also provided.

Description

[Technical field]

[0001] The present invention relates generally to the field of protein constructs, and more particularly to the field of protein constructs comprising tandemly linked IgA Fc and IgG Fc regions. The present invention also relates to such protein constructs further comprising a targeting domain or a targeting ligand, e.g., an antigen binding domain. The present invention further relates to compositions comprising such protein constructs, methods for producing such protein constructs, and therapeutic methods and uses employing such protein constructs.

[0002] Monoclonal antibodies are being applied to treat an increasingly wide range of diseases, with 2020 seeing the second-highest approval rate ever, with cancer-targeting antibodies accounting for the majority of them.

[0003] For example, immunoglobulin (Ig)G-based therapeutics, such as IgG antibodies, can eliminate cancer cells through Fc-mediated effector mechanisms, including classical Fcγ receptor (FcγR) engagement followed by induction of effector functions such as antibody-dependent cellular cytotoxicity (ADCC) and / or antibody-dependent cellular phagocytosis (ADCP). Furthermore, IgG antibodies can recruit the complement system upon target binding and subsequently induce complement-dependent cytotoxicity (CDC).

[0004] This motivated the development of Fc engineering strategies to generate IgG variants with enhanced effector functions. An additional advantage of using IgG antibodies is their long plasma half-life, thus ensuring bioavailability by being rescued from intracellular degradation by the fetal Fc receptor (FcRn), which acts as a homeostatic regulator of IgG catabolism. Plasma half-life can be further extended by Fc engineering to improve FcRn binding.

[0005] However, there is a need for more potent formats that are tailored to both optimal effector function properties and good pharmacokinetics.

[0006] The present invention provides an antibody format that combines structural elements of IgA and IgG and is engineered to efficiently engage both FcγR and FcαR. Specifically, it has been surprisingly found that an IgA Fc region linked to an IgG Fc region, with the C-terminus of the IgA Fc region linked to the N-terminus of the IgG Fc region, exhibits advantageous effector functions as well as efficient engagement of multiple effector molecules. In particular, the constructs of the present invention have improved binding to C1q, a key step in complement activation. In addition, the protein constructs of the present invention exhibit good plasma half-life due to effective FcRn binding and are therefore efficiently rescued from intracellular degradation.

[0007] Furthermore, the present invention surprisingly demonstrates that protein constructs provided by the present invention in which the C-terminus of an IgA Fc region is linked to the N-terminus of an IgG Fc region have improved properties compared to constructs engineered in the opposite orientation, i.e., in which the C-terminus of an IgG Fc region is linked to the N-terminus of an IgA Fc region.

[0008] Thus, in a first aspect, the present invention provides a protein construct comprising an IgA Fc region, preferably a human IgA Fc region, and an IgG Fc region, preferably a human IgG Fc region, wherein the C-terminus of the human IgA Fc region is linked to the N-terminus of the human IgG Fc region.

[0009] As used herein, the term Fc region (or Fc fragment or fragment crystallizable region) includes or corresponds to a portion of an antibody that has the ability to interact with an Fc receptor or that provides (or confers) an antibody effector function (e.g., antibody-dependent cellular cytotoxicity). A naturally occurring Fc region (or Fc fragment) is composed of two identical chains (dimers) that contain (or comprise or consist of) the amino acid sequences of the CH2 and CH3 domains of an antibody. The two chains of the Fc fragment are generally linked to each other by at least one cysteine ​​bridge (a disulfide bond between cysteine ​​residues).

[0010] Truncated, mutated or altered Fc regions (or Fc fragments), e.g., fragmented or variant Fc regions, particularly fragments or variants of IgG-Fc regions or IgA-Fc regions, may also be used, provided that, e.g., the ability to interact with an Fc receptor, e.g., FcRn, and / or to confer effector function is maintained, or present, or improved, as compared to the starting unmutated or wild-type Fc region. By way of example, suitable variants with improved binding to FcR, e.g., FcRn, or with (or conferring) improved or increased half-life or improved or increased effector function, are well known and described in the art, and any of these may be used.

[0011] Thus, the protein construct of the present invention comprises an Fc region (or an Fc fragment) of an IgA antibody. Such an IgA Fc region (or fragment crystallizable region) is capable of binding to the FcαRI receptor and / or of effecting (or conferring) the appropriate IgA antibody effector function (e.g., antibody-dependent cellular cytotoxicity (ADCC) or antibody-dependent cellular phagocytosis (ADCP). As outlined above, such an Fc region typically comprises the CH2 and CH3 domains of an IgA antibody.

[0012] In some embodiments, the IgA Fc region used in the protein constructs of the present invention has an altered (or mutated, inactivated, or truncated) tailpiece or does not include a tail (or the tail is removed).

[0013] Thus, IgA Fc regions (or Fc fragments) include those in which the tail has been modified, mutated, inactivated, truncated, or removed. However, such IgA Fc regions still contain the CH2 and CH3 domains of the IgA antibody. Preferred IgA Fc regions further contain all or a portion of the IgA hinge region, e.g., the portion involved in the formation of the disulfide bond linking the two polypeptide chains comprising at least the IgA Fc region and / or the portion involved in the Fc effector function.

[0014] The tail is typically an 18 amino acid region (or extension) at the C-terminal end of the IgA heavy chain constant region or Fc region that contains the cysteine ​​residues essential for polymerization (e.g., dimer formation).

[0015] Thus, in a preferred embodiment, the IgA tail is modified, mutated, inactivated, truncated or removed in the protein construct of the invention. In some such embodiments, a cysteine ​​residue essential for polymerization is modified, mutated, inactivated, truncated or removed. Such removal or modification of the tail is also advantageous in allowing the attachment of the C-terminus of the IgA Fc region to the N-terminus of the IgG Fc region.

[0016] IgA tails are well known to those skilled in the art (e.g., Janeway et al., 2001, Immunobiology: The Immune System in Health and Disease, 5th Edition, New York: Garland Science). An example of an IgA1 tail can be represented by the sequence PTHVNVSVMAEVDGTCY (SEQ ID NO: 30).

[0017] Typically, there is at least one bond (preferably at least one disulfide bond) between the two polypeptide chains comprising the IgA Fc region, for example, there is typically a disulfide bond between the heavy chains of an IgA antibody.

[0018] The IgA Fc region used in the constructs of the present invention may be from any subtype of IgA antibody, for example from IgA1 or from IgA2, preferably from IgA2.

[0019] Thus, the present invention provides a protein construct wherein the IgA Fc region is an IgA1 Fc region or an IgA2 Fc region, preferably an IgA2 Fc region.

[0020] In some embodiments, the IgA Fc region may be engineered or modified to contain enhanced or altered properties, such as enhanced or altered effector function, which may include induction of antibody-dependent cellular cytotoxicity (ADCC) or antibody-dependent cellular phagocytosis (ADCP), or increased co-engagement or binding to Fcα receptors, such as FcRαI. Suitable Fc variants having (or conferring) these characteristics are well known and described in the art, and any of these may be used.

[0021] As mentioned above, a suitable Fc region (IgA Fc region) for use in the construct of the present invention comprises a CH2 domain and a CH3 domain. In some embodiments, a CH1 domain can be included. However, in other embodiments, only the CH2 domain and the CH3 domain (or a fragment or variant thereof that retains or has the ability to interact with FcαR, e.g., FcαR1, or that provides or confers IgA antibody effector function and preferably dimerization ability) are the portions of the IgA antibody that are included in the construct. For example, in some embodiments, the light chain antibody domain, e.g., the light chain antibody domain of an IgA antibody, in particular the light chain constant domain (CL domain) or the light chain variable domain (VL), will not be included in the construct. In other examples, the heavy chain CH1 domain or the heavy chain variable domain (VH), e.g., the heavy chain CH1 domain or the heavy chain variable domain (VH) of an IgA antibody, will not be included in the construct. In other embodiments, the construct will include a hinge region, such as an IgA hinge region or an IgG hinge region, as described elsewhere herein. If hinge regions are included, they can be conveniently included in their native position, i.e., linked or attached to the N-terminus of the IgA CH2 region, for example, between the IgA CH2 region and the CH1 region (if such a CH1 region is present). Alternatively, such hinge regions can be used to link the IgA CH2 domain of the construct to a targeting domain, as described elsewhere herein.

[0022] However, in some embodiments, for example those in which the IgA Fc region is part of an IgA antibody, a light chain constant domain (CL domain) and / or a light chain variable region (VL), and / or a heavy chain constant domain (CH1 domain) and / or a heavy chain variable region (VH) will be included. In other words, in some embodiments, an antigen binding domain will be present in the construct that comprises a VL domain and / or a VH domain, optionally with a CL domain and / or a CH1 domain. In a preferred embodiment, such an antigen binding domain will be derived from an IgA antibody. In other preferred embodiments, such an antigen binding domain will be derived from an IgG antibody. In a preferred embodiment, a hinge region, for example an IgA hinge region or an IgG hinge region, can be used to link said antigen binding domain, for example an IgA CH2 region, to the CH1 domain.

[0023] In a preferred embodiment, the IgA Fc region is a human IgA Fc region. The sequences encoding the constant region of IgA antibodies and IgA Fc regions, as well as the locations of CH1 domain, hinge domain, CH2 domain, and CH3 domain, are readily available to those skilled in the art. For example, the gene encoding human immunoglobulin heavy constant alpha 1 is known as IGHA1 (see, for example, Uniprot: P01876), and the gene encoding human immunoglobulin heavy constant alpha 2 is known as IGHA2 (see, for example, Uniprot: P01877).

[0024] For example, an exemplary IgA1 heavy chain constant region amino acid sequence is set forth herein as SEQ ID NO:5, and an exemplary IgA2 heavy chain constant region amino acid sequence is set forth herein as SEQ ID NO:6, from which exemplary sequences of the CH1 domain, CH2 domain, and / or CH3 domain, optionally with the hinge region, can be derived for inclusion in the construct. Exemplary sequences are shown in Table 1.

[0025] In some embodiments of the protein constructs of the invention comprising an IgA Fc region, the last (C-terminal most) amino acid residue of the IgA Fc region is the residue corresponding to position 452 in the Bur numbering (e.g., A at position 452 in the Bur numbering). See examples of various sequences shown in Table 1. The Bur numbering scheme is described in Liu et al. (Science. 1976 Sep 10;193(4257):1017-20) and is the preferred numbering system for the IgA molecules and IgA fragments described herein.

[0026] The protein construct of the present invention further comprises an Fc region (or an Fc fragment) of an IgG antibody. Such an IgG Fc region (or fragment crystallizable region) is capable of binding to an IgG Fcγ receptor or FcRn or C1q and / or can provide (or confer) an appropriate IgG antibody effector function (e.g. ADCC or ADCP, or complement dependent cytotoxicity CDC). As outlined above, such an Fc region typically comprises the CH2 and CH3 domains of an IgG antibody.

[0027] Typically, at least one bond (preferably at least one disulfide bond) exists between two polypeptide chains comprising an IgG Fc region. For example, a disulfide bond typically exists between IgG antibody heavy chains. A preferred IgG Fc region further comprises all or a portion of an IgG hinge region, e.g., further comprising a portion involved in the formation of a disulfide bond linking at least two polypeptide chains comprising an IgG Fc region and / or a portion involved in an Fc effector function.

[0028] The IgG Fc region used in the construct of the present invention can be from any subtype of IgG antibody, for example, IgG1, IgG2, IgG3, or IgG4. In some embodiments, an IgG1 Fc region or an IgG2 Fc region is used. In some embodiments, an IgG1 Fc is preferably used. In some embodiments, an IgG2 Fc is preferably used.

[0029] In one particular embodiment, a protein construct comprising a human IgA2 Fc region and a human IgG2 Fc region, wherein the C-terminus of the human IgA2 Fc region is linked to the N-terminus of the human IgG2 Fc region, wherein the human IgG2 Fc region is (i) an arginine (R) residue at position 311 according to EU numbering; (ii) a glutamic acid (E) residue at position 428 according to the EU numbering system; and (iii) a tryptophan (W) residue at position 434 according to EU numbering; wherein an IgG hinge is located between said IgA2 Fc region and said IgG2 Fc region.

[0030] In one particular embodiment, a protein construct comprising a human IgA2 Fc region and a human IgG2 Fc region, wherein the C-terminus of the human IgA2 Fc region is linked to the N-terminus of the human IgG2 Fc region, wherein the human IgG2 Fc region is (i) an arginine (R) residue at position 311 according to EU numbering; (ii) a glutamic acid (E) residue at position 428 according to the EU numbering system; and (iii) a tryptophan (W) residue at position 434 according to EU numbering; wherein a human IgG2 hinge is located between said IgA2 Fc region and said IgG2 Fc region.

[0031] In one particular embodiment, a protein construct comprising a human IgA2 Fc region and a human IgG2 Fc region, wherein the C-terminus of the human IgA2 Fc region is linked to the N-terminus of the human IgG2 Fc region, wherein the human IgG2 Fc region is (i) an arginine (R) residue at position 311 according to EU numbering; (ii) an aspartic acid (D) residue at position 428 according to the EU numbering system; and (iii) a tryptophan (W) residue at position 434 according to EU numbering; wherein an IgG hinge is located between said IgA2 Fc region and said IgG2 Fc region.

[0032] In one particular embodiment, a protein construct comprising a human IgA2 Fc region and a human IgG2 Fc region, wherein the C-terminus of the human IgA2 Fc region is linked to the N-terminus of the human IgG2 Fc region, wherein the human IgG2 Fc region is (i) a lysine (K) residue at position 311 according to EU numbering; (ii) an aspartic acid (D) residue at position 428 according to the EU numbering system; and (iii) a tryptophan (W) residue at position 434 according to EU numbering; wherein an IgG hinge is located between said IgA2 Fc region and said IgG2 Fc region.

[0033] In one particular embodiment, a protein construct comprising a human IgA2 Fc region and a human IgG3 Fc region, wherein the C-terminus of the human IgA2 Fc region is linked to the N-terminus of the human IgG3 Fc region, wherein the human IgG3 Fc region is (i) an arginine (R) residue at position 311 according to EU numbering; (ii) a glutamic acid (E) residue at position 428 according to EU numbering; (iii) a tryptophan (W) residue at position 434 according to the EU numbering system; and (iv) a histidine (H) residue at position 435 according to EU numbering; wherein an IgG hinge is located between said IgA2 Fc region and said IgG3 Fc region.

[0034] In all of the specific embodiments described above, an IgG1 Fc region may be used in place of the IgG2 or IgG3 region with the corresponding mutations.

[0035] Thus, the present invention provides a protein construct, wherein said IgG Fc region is an IgG1 Fc region or an IgG2 Fc region, preferably an IgG1 Fc region.

[0036] In some embodiments, the IgG Fc region may be engineered or modified to include enhanced or altered properties, such as enhanced or altered effector function, which may include antibody-dependent cellular cytotoxicity (ADCC) or antibody-dependent cellular phagocytosis (ADCP), complement-dependent cytotoxicity (CDC), or increased half-life or increased co-engagement or binding to Fcγ receptors.

[0037] As mentioned above, a suitable Fc region (IgG Fc region) for use in the construct of the invention comprises a CH2 domain and a CH3 domain. A CH1 domain will generally not be included as part of an IgG Fc region. However, in some embodiments, a CH1 domain, e.g., an IgG CH1 domain, may be included in the construct of the invention, e.g., located N-terminal to the CH2 and CH3 domains of the IgG Fc region, e.g., located between the IgA Fc region and the IgG Fc region of the construct. However, in other embodiments, only the CH2 and CH3 domains (or fragments or variants thereof that retain or have the ability to interact with FcgR or FcRn or C1q, or fragments or variants thereof that provide or confer IgG antibody effector functions and preferably dimerization ability) are the portions of the IgG antibody that are included in the construct. In other embodiments, a hinge region, e.g., an IgG hinge region, as described elsewhere herein, will be included in the construct. If hinge regions are included, they can be advantageously included in situ, i.e., in the tandem construct of the invention, for example, between the IgG CH2 and IgA CH3 regions, linked or attached to the N-terminus of the IgG CH2 region. Thus, preferably, such hinge regions can be used to link the IgG CH2 domain of the construct to the IgA CH3 domain of the construct. In other words, they can be used to link the IgG Fc region to the IgA Fc region.

[0038] However, in some embodiments, for example those in which an IgA Fc region is joined to the antigen binding domain, a light chain constant domain (CL domain) and / or a light chain variable region (VL), and / or a heavy chain constant domain (CH1 domain) and / or a heavy chain variable region (VH) may be included. In other words, in some embodiments there will be an antigen binding domain in the construct that comprises a VL domain and / or a VH domain, optionally together with a CL domain and / or a CH1 domain. In some embodiments, such an antigen binding domain may originate from an IgG antibody.

[0039] In a preferred embodiment, the IgG Fc region is a human IgG Fc region. The sequences encoding the constant regions of IgG antibodies and IgG Fc regions, as well as the locations of the CH1 domain, hinge domain, CH2 domain, and CH3 domain, are readily available to those skilled in the art. For example, the gene encoding human immunoglobulin heavy constant gamma 1 is known as IGHG1 (see, for example, Uniprot: P01857), the gene encoding human immunoglobulin heavy constant gamma 2 is known as IGHG2 (see, for example, Uniprot: P01859), the gene encoding human immunoglobulin heavy constant gamma 3 is known as IGHG3 (see, for example, Uniprot: P01860), and the gene encoding human immunoglobulin heavy constant gamma 4 is known as IGHG4 (see, for example, Uniprot: P01861).

[0040] For example, an exemplary IgG1 heavy chain constant region amino acid sequence is set forth herein as SEQ ID NO: 1, an exemplary IgG2 heavy chain constant region amino acid sequence is set forth herein as SEQ ID NO: 2, an exemplary IgG3 heavy chain constant region amino acid sequence is set forth herein as SEQ ID NO: 3, and an exemplary IgG4 heavy chain constant region amino acid sequence is set forth herein as SEQ ID NO: 4, from which exemplary sequences of CH1 domain, CH2 domain, and / or CH3 domain can be derived as appropriate, preferably exemplary sequences of CH2 domain and CH3 domain, optionally together with the hinge region, for inclusion in the construct. Exemplary sequences are shown in Table 1. Engineered IgG heavy chain constant regions and IgG Fc variants with altered glycosylation patterns and / or stability are also well known. The well-known EU numbering scheme is the preferred numbering system for the IgG molecules and IgG fragments described herein.

[0041] In the construct of the present invention, the orientation of the Fc region is important. For this reason, the IgA Fc region is located at the N-terminus of the IgG Fc region. Specifically, the C-terminus of the IgA Fc region is bound to the N-terminus of the IgG Fc region, i.e., the C-terminus or C-terminal end of each of the two polypeptides constituting the IgA Fc region is bound to the N-terminus or N-terminal end of each of the two polypeptides constituting the IgG Fc region. The C-terminus of the IgA Fc region can be bound to the N-terminus of the IgG Fc region directly, i.e., without a linker. However, the C-terminus of the IgA Fc region can be bound to the N-terminus of the IgG Fc region indirectly, i.e., by using a linker as described below.

[0042] In the constructs of the invention, both the IgA Fc region and the IgG Fc region are arranged such that the functionality of each Fc region is at least maintained, and preferably enhanced. Suitable molecules and constructs for comparison will be known to those skilled in the art. By way of example, the maintenance or enhancement can be determined, for example, by comparison with the IgA Fc region alone or the IgG Fc region alone in a suitable full-length IgG or full-length IgA antibody (see, for example, Figures 1a or 1b), or with a protein construct that is equivalent to the protein construct of the invention except that the orientation of the IgA and IgG Fc regions is reversed, i.e., a construct in which the C-terminus of the IgG Fc region is linked to the N-terminus of the IgA Fc region. An example of such a molecule is shown in Figure 1c.

[0043] Thus, the constructs of the present invention are capable of binding to an appropriate FcR, such as FcαR or FcγR, or FcRn, and / or capable of Fc effector functions, such as one or more Fc effector functions selected from ADCC induction, CDC induction, and ADCP induction. As is well known and described in the art, ADCC induction and ADCP induction occur through the binding of an appropriate Fc region to an appropriate FcR on various types of cells capable of such functions, such as on macrophages in the case of ADCP, and on PMN cells or NK cells for ADCC. CDC induction generally occurs through the binding of the Fc region to a component of the C1 complex, preferably C1q. The ability to bind to FcRn is important for plasma half-life.

[0044] Thus, the construct of the present invention is capable of binding to FcR or Fc effector function. For example, the construct is capable of binding to various FcαRs, such as Fcα1, various FcγRs, such as one or more of FcγRI, FcγRIIa, FcγRIIb, FcγRIIc, FcγRIIIa, or FcγRIIIb, and / or FcRn. Alternatively or additionally, the construct is capable of CDC induction, ADCC induction, and / or ADCP induction. To induce CDC, the preferred construct of the present invention is capable of binding to C1q.

[0045] Preferably, the protein construct of the present invention comprising an IgA Fc region (or Fc fragment) has or retains the ability to bind to FcαR (e.g., human FcαR). FcαR is also referred to as FcαRI or CD89. In some embodiments, the protein construct of the present invention has the ability to bind to a recombinant version of FcαRI (e.g., recombinant human FcαRI). For example, an IgA Fc region in a protein construct of the present invention that binds to (or crosslinks with) FcαRI expressed on polymorphonuclear leukocytes (PMNs) can result in antibody-dependent cellular cytotoxicity (ADCC). Thus, it is preferred that the inclusion of both an IgA Fc region and an IgG Fc region in a construct of the present invention does not significantly affect the ability of the construct to bind to FcαRI. The present inventors report herein that the inclusion of both IgA and IgG Fc regions in the constructs of the invention does not significantly affect or reduce the ability of the constructs to bind to FcαRI (see Example 1). In other words, the constructs of the invention are capable of binding (or remaining bound to) FcαRI.

[0046] Thus, in some embodiments, the protein construct of the invention comprising an IgA Fc region has a binding ability to FcαRI that is not significantly (or unchanged), or is substantially the same (or is the same or comparable), or is improved, compared to the binding ability to FcαRI of a control protein, said control protein being, for example, a control protein comprising an IgA Fc region, for example an IgA antibody, for example a full-length IgA antibody (for example a wild-type full-length IgA antibody) of a suitable subtype, such as an IgA1 antibody or an IgA2 antibody (see FIG. 1a), selected depending on the subtype of the IgA Fc region contained in the construct of the invention, or another suitable construct comprising a single IgA Fc region, for example a single IgA Fc region as the only Fc region in the construct. Preferably, the control protein is substantially (or identical) or equivalent to the protein construct of the invention to which it is compared, with the exception (or difference) that the control protein construct does not comprise an IgG Fc region as defined in the invention. In some embodiments, the protein construct of the invention comprises an IgA antibody, while an exemplary control protein comprises the same IgA antibody, with the only difference (compared to the protein of the invention) being that the control protein does not comprise an IgG Fc region as defined in the invention.

[0047] A suitable control protein preferably comprises the same or an equivalent targeting domain or the same or an equivalent antigen binding domain as the protein construct of the invention.

[0048] A level of binding to FcαRI that is functionally effective, e.g., that allows Fc effector function, e.g., ADCC, is desired. The ability of the protein construct of the present invention to bind to FcαRI may be evaluated by any suitable method or assay, and the skilled artisan is familiar with suitable assays. The above discussion may concern FcαRI binding determined (or evaluated) by any suitable assay. Typically and preferably, an ELISA assay is used. A particularly preferred assay for determining FcαRI binding of the protein construct of the present invention is described in Example 1 herein.

[0049] The protein construct of the present invention comprising an IgG Fc region (or Fc fragment) preferably has or retains the ability to bind to FcγR (e.g., human FcγR). There are three classes of human IgG receptors found on leukocytes: FcγRI (CD64), FcγRIIa, FcγRIIb, FcγRIIc (CD32), and FcγRIIIa, FcγRIIIb (CD16). The protein construct of the present invention has the ability to bind to one or more, or preferably all, of these Fcγ receptors. In some embodiments, the protein construct of the present invention has the ability to bind to a recombinant version of FcγR (e.g., recombinant human FcγR). For example, the IgG Fc region in the protein construct of the present invention that binds to (or crosslinks) FcγR expressed on effector cells such as natural killer cells, macrophages, monocytes, or eosinophils can result in antibody-dependent cellular cytotoxicity (ADCC). Thus, it is preferred that the inclusion of both IgA and IgG Fc regions in the construct of the present invention does not significantly affect the ability of the construct to bind to FcγR.

[0050] The present inventors report herein that the inclusion of both IgA and IgG Fc regions in the construct of the present invention does not significantly reduce the binding ability of the construct to FcγR, for example, compared to a full-length IgG antibody of an appropriate subtype (e.g., selected according to the subtype of the IgG Fc region included in the construct of the present invention) (see Examples 1 and 2). Binding to each of the three classes of FcγR has been tested, with the construct of the present invention showing good binding to one or more of them. Specifically, binding to FcγRI, FcγRIIa (both R131 and H131 variants), FcγRIIb, FcγRIIIa (both F158 and V158 variants), and FcγRIIIb has been tested. In each case, binding to FcγR is at least equivalent to that of the appropriate full-length IgG antibody. Surprisingly, in some instances, the binding of the constructs of the invention to FcγRI, FcγRIIa (both the R131 and H131 variants), FcγRIIb, FcγRIIIa-V, and FcγRIIIb in particular is improved (increased) or superior to that of full-length IgG antibodies. Moreover, in some instances, the binding of the constructs of the invention is improved (increased) or superior to that of constructs with the opposite orientation to that of the constructs of the invention, i.e., constructs in which the C-terminus of the IgG Fc region is linked to the N-terminus of the IgA Fc region.

[0051] NK cells express FcγRIIIa and are believed to be the major cell population mediating ADCC upon cross-linking of IgG antibodies. Therefore, the improved FcγRIIIa binding ability of the constructs of the present invention is considered to be particularly advantageous in that it allows for improved binding to NK cells, thereby efficiently mediating ADCC.

[0052] Thus, in some embodiments, the protein construct of the invention comprising an IgG Fc region has a binding ability to FcγR that is not significantly (or unchanged), or is substantially (or is equivalent or comparable), or is improved (or increased) compared to the binding ability of a control protein to an equivalent FcγR, said control protein being, for example, a control protein comprising an IgG Fc region, for example an IgG antibody, for example a full-length IgG antibody (for example a wild-type full-length IgG antibody) of an appropriate subtype, such as an IgG1 antibody, an IgG2 antibody, an IgG3 antibody, or an IgG4 antibody, or another suitable construct comprising a single IgG Fc region, for example a single IgG Fc region as the only Fc region in said construct. Preferably, the control protein is substantially (or identical) or equivalent to the protein construct of the invention to which it is compared, with the exception (or difference) that said control protein construct does not comprise an IgA Fc region as defined in the invention.

[0053] A suitable control protein preferably comprises the same or an equivalent targeting domain or the same or an equivalent antigen binding domain as the protein construct of the invention.

[0054] It is desirable to bind to FcγR at a level that is functionally effective, e.g., at any of the above-mentioned FcγRs, e.g., at a level that allows Fc effector function, e.g., ADCC. The ability of the protein construct of the present invention to bind to FcγR may be evaluated by any suitable method or assay, and the skilled person is familiar with suitable assays. The above discussion may be in terms of FcγR binding determined (or evaluated) by any suitable assay. Typically and preferably, an ELISA assay is used. A particularly preferred assay for determining FcγR binding of the protein construct of the present invention is described in Example 1 herein.

[0055] FcRn is a type I membrane glycoprotein that is expressed primarily in acidic intracellular compartments such as endosomes. One of the known roles of FcRn is in recycling certain molecules, such as IgG or albumin, back to serum after endocytosis. For example, FcRn interacts with the Fc region of IgG at the interface of the CH2 and CH3 domains in a 2:1 stoichiometry (i.e., one molecule of IgG-Fc binds two molecules of FcRn). Recycling is facilitated by pH-dependent binding of IgG-Fc to FcRn. In this regard, IgG-Fc binds to FcRn with high affinity at pH 6.0 or 6.5, but not at pH 7.4. Thus, FcRn binds IgG (via the IgG-Fc region) in acidified endosomes, but subsequent dissociation of IgG from FcRn at physiological / neutral pH occurs, for example, when the recycling endosome containing the FcRn-IgG complex fuses with the cell membrane to release IgG back into serum. This mechanism allows IgG to avoid lysosomal degradation, and recycling via FcRn is thought to be the cause of the long half-life of IgG in circulation (serum half-lives of IgG1, IgG2, and IgG4 are about 21 days, while that of IgG3, which has a lower binding efficiency to FcRn, is about 7 days). The Fc region of IgA antibodies does not bind well to FcRn, which is thought to be one of the reasons for their relatively short serum half-life (about 1 day).

[0056] Preferably, the protein construct of the present invention comprising an IgG Fc region (or Fc fragment) has or retains the ability to bind to FcRn (e.g. human FcRn), preferably in a pH-dependent manner. In some embodiments, the protein construct of the present invention has the ability to bind to a recombinant version of FcRn (e.g. recombinant human FcRn). As described elsewhere herein, the IgG Fc region binds to FcRn, which means that IgG antibodies are rescued from degradation by a pH-dependent mechanism and have a relatively long serum half-life. Thus, it is preferred that the inclusion of both an IgA Fc region and an IgG Fc region in the construct of the present invention does not significantly affect the pH-dependent binding ability of the construct to FcRn. The present inventors report herein that the inclusion of both an IgA Fc region and an IgG Fc region in a construct of the invention does not significantly affect or reduce the ability of the construct to bind to FcRn, e.g., compared to a full-length IgG antibody of an appropriate subtype (e.g., selected depending on the subtype of the IgG Fc region included in the construct of the invention) (see Example 1).

[0057] Thus, in some embodiments, the protein construct of the present invention has a binding ability to FcRn, preferably pH-dependent. More specifically, the protein construct of the present invention preferably has a binding ability to FcRn at acidic pH (e.g., pH 5.5), but has no significant binding ability (or binding ability) to FcRn at neutral pH (e.g., pH 7.4) (or shows no significant binding or binding, or shows significantly reduced binding, or shows weak or low affinity binding). Surprisingly, the inventors have shown that the binding ability of the construct of the present invention to FcRn is improved (or increased) or superior to that of a full-length IgG antibody (see Example 1). It has also been shown that the binding of the construct of the present invention to FcRn is improved (or increased) or superior to that of a construct with an orientation opposite to that of the construct of the present invention, i.e., a construct in which the C-terminus of the IgG Fc region is linked to the N-terminus of the IgA Fc region. In some embodiments, FcRn binding can be further improved by including a REW mutation as described elsewhere herein.

[0058] Thus, in some embodiments, the binding ability of the protein construct of the present invention to FcRn, preferably the pH-dependent binding ability to FcRn, is not significantly changed (or unchanged), or is substantially equivalent (or equivalent or comparable), or is improved (increased) compared to the binding ability to FcRn of a control protein, said control protein being, for example, a control protein comprising an IgG Fc region, for example an IgG antibody, for example a full-length IgG antibody of an appropriate subtype, such as an IgG1 antibody, an IgG2 antibody, an IgG3 antibody, or an IgG4 antibody (for example a wild-type full-length IgG antibody), or another suitable construct comprising a single IgG Fc region, for example a single IgG Fc region as the only Fc region in said construct. Preferably, the control protein is substantially identical (or identical) or equivalent to the protein construct of the present invention to which it is compared, with the exception (or difference) being that said control protein construct does not comprise an IgA Fc region as defined in the present invention. A suitable control protein preferably comprises the same or an equivalent targeting domain or the same or an equivalent antigen binding domain as the protein construct of the invention.

[0059] Binding to FcRn at a level that is functionally effective, e.g., that allows recycling of the construct and / or that allows a comparable or improved serum half-life, is desired. The ability of the protein construct of the present invention to bind to FcRn may be evaluated by any suitable method or assay, and the skilled artisan is familiar with suitable assays. The above discussion may relate to FcRn binding determined (or evaluated) by any suitable assay. Typically and preferably, an ELISA assay is used. Typically and preferably, to evaluate the pH dependency of FcRn binding, parallel ELISA assays may be performed, with an assay at neutral pH (e.g., pH 7.4) and an assay at acidic pH (e.g., pH 5.5). A particularly preferred assay for determining FcRn binding of the protein construct of the present invention is described in Example 1 herein.

[0060] Preferably, the protein constructs of the invention can be recycled in an FcRn-dependent (or FcRn-mediated) manner and rescued from intracellular degradation (e.g., from lysosomal degradation). As described elsewhere herein, IgA antibodies typically suffer from a short in vivo half-life (e.g., a half-life of about 1 day). The inventors have demonstrated herein that the inclusion of both an IgA Fc region and an IgG Fc region in the constructs of the invention can confer (or enhance or increase) recycling and rescue capabilities via an FcRn-dependent mechanism to an IgA antibody (or other constructs in which an IgA Fc region is present or in which the IgA Fc region is the only Fc region) (see Example 1).

[0061] In addition, the inventors have demonstrated herein that the inclusion of both IgA Fc and IgG Fc regions in the constructs of the invention does not reduce the recycling and rescue capacity of the constructs containing IgG Fc regions (or Fc fragments) via FcRn-dependent mechanisms (see Example 1). Indeed, the inventors have shown that the recycling and rescue of the constructs of the invention via FcRn is at least equivalent to that of a suitable full-length IgG antibody, e.g., an IgG1 antibody. Surprisingly, the recycling and rescue of the constructs of the invention via FcRn is shown to be improved (increased) or superior to that of a full-length IgG antibody. Also, the recycling and rescue of the constructs of the invention via FcRn is shown to be improved (increased) or superior to that of a construct with an orientation opposite to that in the constructs of the invention, i.e., a construct in which the C-terminus of the IgG Fc region is linked to the N-terminus of the IgA Fc region. Without wishing to be bound by theory, it is believed that this ability to bind to FcRn and be recycled and rescued is responsible for the increased in vivo half-life observed in the protein constructs of the present invention (as described elsewhere herein).

[0062] In a preferred embodiment, the recycling and rescue ability of the protein construct of the present invention via an FcRn-dependent mechanism is maintained or not significantly altered (e.g., compared to an IgG Fc-containing control protein) or is increased (preferably significantly increased) compared to the recycling and rescue ability of a control protein via an FcRn-dependent mechanism.

[0063] In some embodiments, the ability of the protein construct of the invention to recycle and rescue via an FcRn-dependent mechanism is increased, e.g., by at least 1.2-fold, at least 1.3-fold, at least 1.4-fold, at least 1.5-fold, at least 2-fold, at least 3-fold, at least 4-fold, at least 5-fold, at least 6-fold, at least 7-fold, at least 8-fold, at least 9-fold, at least 10-fold, at least 11-fold, at least 12-fold, at least 13-fold, at least 14-fold, or at least 15-fold (e.g., up to 1.5-fold, up to 2-fold, up to 3-fold, up to 4-fold, up to 5x, up to 6x, up to 7x, up to 8x, up to 9x, up to 10x, up to 11x, up to 12x, up to 13x, up to 14x, up to 15x, up to 20x, up to 50x, or up to 100x increase, 1.2x or 1.5x to 100x increase, 1.2x or 1.5x to 50x increase, 1.2x or 1.5x to 20x increase, 1.2x or 1.5x to 10x increase, 3x to 100x increase, 3x to 50x increase, 3x to 20x increase, 3x to 10x increase, 5x to 100x increase, 5x to 50x, 5x to 20x, or 5x to 10x increase, 10x to 100x increase, 10x to 50x, 10x to 20x, or 10x to 15x increase, etc.

[0064] If an increase in recycling and rescue is achieved, a suitable control protein is a control protein comprising an IgA Fc region, for example an IgA antibody, for example a full-length IgA antibody (for example a wild-type full-length IgA antibody) of a suitable subtype, such as an IgA1 antibody or an IgA2 antibody, selected depending on the subtype of the IgA Fc region contained in the construct of the invention, or another suitable construct comprising a single IgA Fc region, for example a single IgA Fc region as the only Fc region in the construct. Preferably, such a control protein is substantially identical (or identical) or equivalent to the protein construct of the invention to which it is compared, with the exception (or difference) being that the control protein construct does not comprise an IgG Fc region as defined in the invention. In some embodiments, the protein construct of the invention comprises an IgA antibody, while an exemplary control protein comprises the same IgA antibody, with the only difference (relative to the protein of the invention) being that the control protein does not comprise an IgG Fc region as defined in the invention.

[0065] Alternatively, if an increase in recycling and rescue is achieved or if maintenance of recycling and rescue is observed, e.g., if recycling and rescue are not significantly altered (or are unchanged) or are substantially equivalent (or equivalent or comparable) to the recycling and rescue abilities of the control protein, then a suitable control protein may be a control protein comprising an IgG Fc region, e.g. an IgG antibody, e.g. a full-length IgG antibody of a suitable subtype such as an IgG1 antibody, an IgG2 antibody, an IgG3 antibody or an IgG4 antibody (e.g. a wild-type full-length IgG antibody), or another suitable construct comprising a single IgG Fc region, e.g. a single IgG Fc region as the only Fc region in the construct. Preferably, the control protein is substantially identical (or identical) or equivalent to the protein construct of the invention to which it is compared, with the exception (or difference) that the control protein construct does not comprise an IgA Fc region as defined in the present invention.

[0066] A suitable control protein preferably comprises the same or an equivalent targeting domain or the same or an equivalent antigen binding domain as the protein construct of the invention. The ability of the protein construct of the present invention to recycle and rescue via an FcRn-dependent mechanism may be determined by any suitable assay or method, which the skilled artisan will be familiar with. The above discussion may relate to the recycling and rescue via an FcRn-dependent mechanism as determined (or assessed) by any suitable assay.

[0067] In some embodiments, the ability of the protein of the present invention to be recycled and rescued via FcRn-dependent mechanism is determined (or evaluated) in a human endothelial cell-based recycling assay (HERA). Such a HERA can also be used to evaluate the pH-dependent binding ability of the protein construct of the present invention to FcRn. Suitable HERAs are known in the art (e.g., Grevys et al., 2018, Nat. Commun., 9(1):621). An exemplary HERA for determining the recycling and rescue ability of the protein construct of the present invention via FcRn-dependent mechanism is described in Example 1 herein.

[0068] Typically and preferably, the protein constructs of the invention have a therapeutically useful in vivo half-life (e.g., a therapeutically useful half-life in the mammalian circulation, preferably in humans, or a therapeutically useful serum half-life or a therapeutically useful plasma half-life). IgA antibodies typically suffer from a short in vivo half-life (e.g., a half-life of about 1 day). The inventors have demonstrated herein that the inclusion of both an IgA Fc region and an IgG Fc region in the constructs of the invention does not significantly reduce the in vivo half-life of the constructs (see Example 1). Indeed, the constructs of the invention have been shown to significantly extend (or increase) the in vivo half-life of IgA-based proteins, such as IgA antibodies (e.g., full-length, e.g., wild-type IgA1 or IgA2 antibodies) or IgA-based antibodies (e.g., antibodies based on IgA1 or IgA2).

[0069] This is an advantageous property in that IgA antibody or protein constructs comprising an IgA Fc region, and in particular IgA antibody or protein constructs comprising an IgA Fc region as the only Fc region, which might otherwise be impossible (or less than optimal for use) in vivo (e.g., for therapeutic purposes) due to their short half-life, can be modified to have an extended half-life by preparing or incorporating them into protein constructs in accordance with the present invention.

[0070] In addition, the inventors have demonstrated herein that the inclusion of both IgA and IgG Fc regions in the constructs of the invention does not significantly reduce the in vivo half-life of the constructs (see Example 1). Indeed, the inventors have shown that the in vivo half-life of the constructs of the invention is not significantly different or is equivalent (or only slightly shorter) than the in vivo half-life of a suitable full-length IgG antibody, e.g., an IgG1 antibody. Furthermore, the in vivo half-life of the constructs of the invention has been shown to be improved (increased) or superior to that of a construct with an orientation opposite to that of the constructs of the invention, i.e., a construct in which the C-terminus of the IgG Fc region is linked to the N-terminus of the IgA Fc region. In some embodiments, the in vivo half-life can be further improved by including a REW mutation as described elsewhere herein.

[0071] In preferred embodiments, the in vivo half-life of the protein construct of the invention is maintained or not significantly altered or is increased (preferably significantly increased) compared to the in vivo half-life of a control protein (e.g. compared to an IgG Fc-containing control protein).

[0072] In some embodiments, the in vivo half-life of the protein construct of the invention is at least 3 days, or at least 4 days, or at least 5 days, or at least 6 days, or at least 7 days (e.g., up to 4 days, up to 5 days, up to 6 days, up to 7 days, up to 8 days, up to 10 days, up to 15 days, up to 20 days, up to 22 days, or up to 25 days, e.g., 3-5 days, 3-7 days, 4-7 days, or 5-7 days, or 6-7 days, or 4-22 days, or 5-22 days, or 6-22 days). In preferred embodiments, the in vivo half-life is the in vivo half-life in a mammal, e.g., the in vivo half-life assessed in an experimental animal such as a mouse, or in a human. The in vivo half-life may be the serum half-life or plasma half-life (or half-life in the circulation or bloodstream) in an experimental animal such as a mouse, particularly a mouse expressing human FcRn rather than mouse FcRn, or in a human. In some embodiments, the in vivo half-life is a beta-phase half-life.

[0073] In some preferred embodiments, the protein constructs of the invention, e.g., proteins including IgA antibodies (e.g., IgA1 or IgA2 antibodies), have an increased (preferably significantly increased) in vivo half-life compared to the in vivo half-life of a control protein.

[0074] In some embodiments, the in vivo half-life of a protein construct of the invention (e.g., a protein comprising an IgA antibody (e.g., an IgA1 antibody or an IgA2 antibody)) is increased by at least 3-fold, at least 4-fold, at least 5-fold, at least 6-fold, at least 7-fold, at least 8-fold, at least 9-fold, or at least 10-fold compared to the in vivo half-life of a control protein (e.g., an increase of up to 3-fold, up to 4-fold, up to 5-fold, up to 6-fold, up to 7-fold, up to 8-fold, up to 9-fold, up to 10-fold, up to 15-fold, or up to 20-fold compared to the in vivo half-life of a control protein, including a 3-fold to 5-fold increase, a 3-fold to 8-fold increase, a 3-fold to 10-fold increase, a 3-fold to 20-fold increase, a 5-fold to 8-fold increase, a 5-fold to 10-fold increase, or a 5-fold to 20-fold increase, etc.).

[0075] If an increase in in vivo half-life is achieved, a suitable control protein is a control protein comprising an IgA Fc region, e.g. an IgA antibody, e.g. a full-length IgA antibody (e.g. a wild-type full-length IgA antibody) of a suitable subtype, such as an IgA1 antibody or an IgA2 antibody, selected depending on the subtype of the IgA Fc region contained in the construct of the invention, or another suitable construct comprising a single IgA Fc region, e.g. a single IgA Fc region as the only Fc region in the construct. Preferably, such a control protein is substantially identical (or identical) or equivalent to the protein construct of the invention to which it is compared, with the exception (or difference) being that the control protein construct does not comprise an IgG Fc region as defined in the invention. In some embodiments, the protein construct of the invention comprises an IgA antibody, while an exemplary control protein comprises the same IgA antibody, with the only difference (relative to the protein of the invention) being that the control protein does not comprise an IgG Fc region as defined in the invention.

[0076] Alternatively, if a maintenance or at least a maintenance of the in vivo half-life is observed, e.g., the in vivo half-life is not significantly altered (or unchanged) or is substantially equivalent (or equivalent or comparable) to that of the control protein, then a suitable control protein may be a control protein comprising an IgG Fc region, e.g., a full-length IgG antibody of a suitable subtype, such as an IgG1, IgG2, IgG3 or IgG4 antibody (e.g., a wild-type full-length IgG antibody), or another suitable construct comprising a single IgG Fc region, e.g., a single IgG Fc region as the only Fc region in the construct. Preferably, the control protein is substantially identical (or identical) or equivalent to the protein construct of the invention to which it is compared, with the exception (or difference) that the control protein construct does not comprise an IgA Fc region as defined in the present invention.

[0077] A suitable control protein preferably comprises the same or an equivalent targeting domain or the same or an equivalent antigen binding domain as the protein construct of the invention. In vivo half-life (or plasma half-life or serum half-life) may be determined by any suitable method, of which those of skill in the art will be familiar. In some embodiments, the in vivo half-life is determined in experimental animals (e.g., experimental mice), and in some embodiments, the considerations and values ​​related to in vivo half-life relate to in vivo half-life determined in mouse models (e.g., as described elsewhere herein). In other embodiments, the considerations and values ​​related to in vivo half-life relate to in vivo half-life observed in humans.

[0078] In such a preferred method for determining in vivo half-life, a mouse, such as a mouse expressing human FcRn but not mouse FcRn (such as homozygous Tg32 mouse (B6.Cg-Fcgrttm1Dcr Tg(FCGRT)32Dcr / DcrJ; The Jackson Laboratory)), is used to determine the half-life of the protein construct of the present invention. A preferred mouse is described in Example 1. An exemplary method for determining in vivo half-life is described in Example 1 herein.

[0079] The inventors have advantageously demonstrated herein that the constructs of the invention also perform well in a mouse model that is more representative of the actual in vivo or physiologically relevant situation (see Example 1). In this mouse model, e.g., a modification of the mouse model described above, mice are pre-loaded with a collection of pooled human IgG molecules, e.g., in the form of an IVIg formulation. The presence of such human IgG molecules in excess is believed to more accurately represent the in vivo situation where such molecules are present, e.g., in the form of endogenous antibodies, and where such high levels of IgG compete for FcRn binding and recycling. Thus, there may be competition for the IgG binding site on FcRn. The inventors have advantageously demonstrated that the constructs of the invention exhibit favorable half-life even in the presence of IgG molecules that compete with each other. Thus, in such models, the in vivo half-life of the constructs of the invention is not significantly altered (or unchanged), or is substantially equivalent (or equivalent or comparable), or only slightly shorter, than the in vivo half-life of a suitable control protein as described elsewhere herein, e.g., a wild-type full-length IgG antibody such as IgG1. This effect has been demonstrated, for example, in human FcRn transgenic mice, as described above and in Example 1. In some embodiments, the in vivo half-life in such models can be further improved by including a REW mutation as described elsewhere herein.

[0080] Another important effector function that some antibodies have is the ability to activate the classical complement pathway (known as complement-dependent cytotoxicity or CDC). The ability of an antibody or other suitable protein construct to induce CDC is generally mediated by the binding of the Fc region of the antibody or construct to a component of the C1 complex, preferably C1q. This interaction then induces lysis of the target cell to which the antibody (or construct) is bound, via CDC. Generally, the ability to induce CDC varies depending on the Fc region of interest and whether (or to what extent) the Fc region can bind to C1q. For example, wild-type human IgG1 Fc regions and wild-type human IgG3 Fc regions generally have good abilities to bind to C1q and induce CDC. IgG2 Fc regions or IgG4 Fc regions generally have weak or low abilities to do this. Human IgA Fc regions generally do not have the ability to bind to C1q and induce CDC, or have weak abilities to bind to C1q and induce CDC.

[0081] Preferably, the protein constructs of the present invention have or retain the ability to bind C1q (eg human C1q) or exhibit increased or improved binding ability to C1q (eg human C1q).

[0082] As a consequence of this, the protein constructs of the invention preferably therefore have or retain CDC inducibility or exhibit increased or improved CDC inducibility.

[0083] In some embodiments, the protein construct of the invention has the ability to bind to a recombinant version of C1q (e.g., recombinant human C1q). As described elsewhere herein, IgG Fc regions, particularly IgG1 Fc regions and IgG3 Fc regions, have good binding to C1q, meaning that antibodies of these serotypes generally exhibit good CDC activity. Thus, in some embodiments, protein constructs comprising IgG1 or IgG3 Fc regions, particularly IgG1 Fc regions, are preferred. However, IgA Fc regions, such as IgA1 Fc regions and IgA2 Fc regions, generally have poor or no CDC activity. Thus, it is preferred that the inclusion of both IgA and IgG Fc regions in the construct of the invention does not significantly affect the ability of the construct to bind to C1q, for example via the IgG Fc region. Furthermore, the inclusion of both an IgA Fc region and an IgG Fc region in a construct of the invention is preferred, as it can confer (or enhance or increase) the ability of an IgA antibody (or other construct in which the IgA Fc region is present or is the only Fc region) to bind C1q.

[0084] The inventors have demonstrated herein that the inclusion of both an IgA Fc region and an IgG Fc region in a construct of the invention can confer (or enhance or increase) the ability of an IgA antibody (and thus other constructs in which an IgA Fc region is present or in which the IgA Fc region is the only Fc region) to bind C1q, e.g., compared to a full-length IgA antibody (e.g. a wild-type full-length IgA antibody) of an appropriate subtype, selected depending on the subtype of the IgA Fc region included in the construct of the invention (see Example 1).

[0085] In addition, the inventors have demonstrated herein that the inclusion of both an IgA Fc region and an IgG Fc region in a construct of the invention does not reduce the ability of the construct comprising an IgG Fc region (or Fc fragment) to bind C1q, for example, compared to a full-length IgG antibody of the appropriate subtype (e.g., a wild-type full-length IgA antibody) (see Example 1). Indeed, the inventors have shown that the ability of the construct of the invention to bind C1q is at least equivalent to the ability of a suitable full-length IgG antibody (e.g., a wild-type full-length IgG antibody), for example an IgG1 antibody or an IgG2 antibody, to bind C1q. Surprisingly and advantageously, the ability of the construct of the invention comprising an IgG1 Fc region to bind C1q has been shown to be improved or superior to the ability of a full-length IgG1 antibody to bind C1q. This improvement observed for the constructs of the invention having an IgG1 Fc region and an IgA Fc region compared to an IgG1 Fc region alone (e.g. in a full-length IgG1 antibody) was unexpected, and this improved ability to bind C1q should also lead to an improved CDC-inducing ability of the protein constructs of the invention. Indeed, this has been demonstrated in the protein constructs of the invention (see Examples 1 and 2), which show comparable or improved or increased CDC-inducing ability compared to, for example, a suitable full-length IgG antibody (e.g. a wild-type full-length IgG antibody), such as an IgG1 antibody or an IgG2 antibody. CDC induction can be further improved by including a REW mutation as described elsewhere herein. In addition, the extended constructs of the invention described elsewhere herein and in Example 2, with or without the REW mutation, also show such an improved CDC-inducing ability.

[0086] It has also been shown that the C1q-binding ability of the construct of the present invention is improved or superior to that of a construct having an orientation opposite to that of the construct of the present invention, i.e., a construct in which the C-terminus of the IgG1 Fc region is linked to the N-terminus of the IgA Fc region. It has also been shown that the CDC-inducing ability of the construct of the present invention is improved (increased) or superior to that of a construct having an orientation opposite to that of the construct of the present invention, i.e., a construct in which the C-terminus of the IgG Fc region is linked to the N-terminus of the IgA Fc region.

[0087] Thus, in a preferred embodiment, the ability of the protein construct of the present invention to bind C1q is maintained or not significantly altered (e.g. compared to an IgG Fc-containing control protein) or is increased (preferably significantly increased) compared to the ability of the control protein to bind C1q.

[0088] Thus, in a preferred embodiment, the CDC-inducing ability of the protein construct of the invention is maintained or not significantly altered (e.g. compared to an IgG Fc-containing control protein) or is increased (preferably significantly increased) compared to the CDC-inducing ability of the control protein.

[0089] In some embodiments, the protein constructs of the invention are capable of inducing CDC of certain cells, e.g., cancer cells, e.g., cells expressing CD20 or HER2, or cancer cells, depending on the targeting domain used. For illustrative purposes only and not intended to be an exhaustive list, the protein constructs of the invention have been shown to be capable of inducing CDC of cancer cells expressing CD20, e.g., cancerous B-cell lines, including Raji cells, WSU-NHL cells, or SU-DH14 cells.

[0090] Induction of CDC can be easily measured by determining the % lysis of the appropriate target cells. The value of % lysis will of course vary depending on the type of cells of interest. However, with the exemplary constructs of the present invention, the % lysis level of cancer cells is good, for example, at least or up to 40% lysis is observed for different types of cancer cells, and at least or up to 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, or 85% lysis levels are observed. For illustrative purposes only, the protein constructs of the present invention with CD20 targeting domain can induce at least or up to 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, or 85% lysis of Raji cells (a CD20-expressing cancer cell line).

[0091] In some embodiments, the ability of the protein construct of the invention to bind to C1q (or to induce CDC) is, for example, at least 1.1-fold, at least 1.2-fold, at least 1.3-fold, at least 1.4-fold, at least 1.5-fold, at least 1.6-fold, at least 1.7-fold, at least 1.8-fold, at least 1.9-fold, at least 2-fold, at least 3-fold, at least 4-fold, at least 5-fold, at least 6-fold, at least 7-fold, at least 8-fold, at least 9-fold, at least 10-fold, at least 12-fold, at least 14-fold, at least 16-fold, at least 18-fold, at least 19-fold, at least 20-fold, at least 25-fold, at least 26-fold, at least 28-fold, at least 29-fold, at least 30-fold, at least 31-fold, at least 32-fold, at least 33-fold, at least 34-fold, at least 35-fold, at least 36-fold, at least 37-fold, at least 38-fold, at least 39-fold, at least 40-fold, at least 41-fold, at least 42-fold, at least 43-fold, at least 45-fold, at least 46-fold, at least 47-fold, at least 48-fold, at least 49-fold, at least 50-fold, at least 51-fold, at least 52-fold, at least 53-fold, at least 54-fold, at least 55-fold, at least 56-fold, at least 57-fold, at least 58-fold, at least 59-fold, at least 59-fold, at least 60-fold, at least 61-fold fold, or at least a 10-fold increase (e.g., up to 1.1-fold, up to 1.2-fold, up to 1.3-fold, up to 1.4-fold, up to 1.5-fold, up to 1.6-fold, up to 1.7-fold, up to 1.8-fold, up to 1.9-fold, up to 2-fold, up to 3-fold, up to 4-fold, up to 5-fold, up to 6-fold, up to 7-fold, up to 8-fold, up to 9-fold, up to 10-fold, up to 15-fold, or up to 20-fold increase, including a 1.2-fold to 20-fold increase, a 1.2-fold to 10-fold increase, a 1.2-fold to 8-fold increase, a 1.3-fold to 20-fold increase, a 1.3-fold to 10-fold increase, a 1.3-fold to 8-fold increase, etc.).

[0092] If an increase in the ability to bind to C1q (or to induce CDC) is achieved, a suitable control protein is a control protein comprising an IgA Fc region, for example an IgA antibody, for example a full-length IgA antibody (for example a wild-type full-length IgA antibody) of a suitable subtype, such as an IgA1 antibody or an IgA2 antibody, selected depending on the subtype of the IgA Fc region contained in the construct of the invention, or another suitable construct comprising a single IgA Fc region, for example a single IgA Fc region as the only Fc region in the construct. Preferably, such a control protein is substantially identical (or identical) or equivalent to the protein construct of the invention to which it is compared, with the exception (or difference) being that the control protein construct does not comprise an IgG Fc region as defined in the invention. In some embodiments, the protein construct of the invention comprises an IgA antibody, while an exemplary control protein comprises the same IgA antibody, with the only difference (compared to the protein of the invention) being that the control protein does not comprise an IgG Fc region as defined in the invention.

[0093] Alternatively, if an increase in the ability to bind to C1q (or to induce CDC) is achieved or if maintenance of the ability to bind to C1q (or to induce CDC) is observed, e.g. if the ability to bind to C1q (or to induce CDC) is not significantly changed (or is unchanged) or is substantially equivalent (or equivalent or comparable) to the ability to bind to C1q (or to induce CDC) of the control protein, a suitable control protein may be a control protein comprising an IgG Fc region, for example an IgG antibody, for example a full-length IgG antibody of a suitable subtype such as an IgG1 antibody, an IgG2 antibody, an IgG3 antibody or an IgG4 antibody, selected depending on the subtype of the IgG Fc region comprised in the construct of the invention (for example a wild-type full-length IgG antibody), or another suitable construct comprising a single IgG Fc region, for example comprising a single IgG Fc region as the only Fc region in the construct. Preferably, such a control protein is substantially identical (or identical) or equivalent to the protein construct of the invention to which it is compared, with the exception (or difference) that the control protein construct does not contain the IgA Fc region as defined in the present invention.

[0094] A suitable control protein preferably comprises the same or an equivalent targeting domain or the same or an equivalent antigen binding domain as the protein construct of the invention.

[0095] A level of binding to C1q that is functionally effective, for example, at a level that allows induction of CDC activity, is desired. The ability of the protein construct of the present invention to bind to C1q may be evaluated by any suitable method or assay, and the skilled person is familiar with suitable assays. The above considerations may relate to C1q binding determined (or evaluated) by any suitable assay. Typically and preferably, an ELISA assay is used. A particularly preferred assay for determining C1q binding of the protein construct of the present invention is described in Example 1 herein. Similarly, the ability of the protein construct of the present invention to induce CDC may be evaluated by any suitable method or assay, and the skilled person is familiar with suitable assays. The above considerations may relate to CDC induction ability determined (or evaluated) by any suitable assay. Typically and preferably, an assay is used that allows measurement of CDC-induced lysis of target cells (i.e., cells expressing a target antigen recognized by the construct of the present invention are used), for example, a fluorescence release assay such as a chromium release assay or a calcein AM-based CDC assay is used. Examples of suitable cells, e.g., cancer cells, e.g., cancerous B cells, are as described in the Examples herein and include Raji cells, WSU-NHL cells, or SU-DH14 cells. A particularly preferred assay for determining the CDC activity of the protein constructs of the present invention is described in Example 1 herein.

[0096] Another important effector function of some antibodies is their ability to induce ADCC. The ability of an antibody or other suitable protein construct to induce ADCC is generally mediated by the binding of the Fc region of the antibody or construct to the appropriate FcR on the effector cell. This interaction can then trigger a signaling cascade in the effector cell, resulting in the secretion of various factors and ultimately the destruction of the target cell to which the antibody (or construct) is bound. In general, the ability to induce ADCC and the efficacy of ADCC induction vary depending on the Fc region of interest and whether (or to what extent) the Fc region can bind to FcR on the effector cell. Induction of ADCC also depends on the ratio of activating and inhibitory receptors. For example, IgG antibodies can induce ADCC via binding to FcγR on effector cells such as natural killer (NK) cells, macrophages, monocytes, and eosinophils). Wild-type human IgG1 Fc regions and wild-type human IgG3 Fc regions generally have excellent (high) ADCC induction ability, while IgG2 Fc regions or IgG4 Fc regions generally have weaker or lower ADCC induction ability. On the other hand, IgA antibodies (e.g., IgA1 antibodies and IgA2 antibodies) generally have good (high) ADCC induction ability through binding to FcαR (FcαRI) on polymorphonuclear leukocytes (PMN), mainly on neutrophils.

[0097] Thus, in a preferred embodiment, the protein construct of the present invention has or retains antibody-dependent cellular cytotoxicity (ADCC) induction ability. Thus, in a preferred embodiment, the ADCC induction ability of the protein construct of the present invention is maintained, or is not significantly changed, or is not significantly decreased (e.g., compared to an appropriate Fc-containing control protein, e.g., a full-length IgA antibody or a full-length IgG antibody (e.g., a wild-type full-length IgA antibody or a wild-type full-length IgA antibody) of a subtype selected according to the subtype of the IgA Fc region and / or IgG Fc region contained in the construct of the present invention), or the ADCC induction ability is increased (preferably significantly increased) compared to the ADCC induction ability of, for example, an appropriate Fc-containing control protein (e.g., a full-length IgA or full-length IgG antibody, e.g., a wild-type full-length IgA antibody or a wild-type full-length IgG antibody of a suitable subtype selected according to the subtype of the IgA Fc region and / or IgG Fc region contained in the construct of the present invention).

[0098] For constructs of the present invention comprising an IgA Fc region, the construct is preferably capable of inducing ADCC via the participation of neutrophils. For constructs of the present invention comprising an IgG Fc region, the construct is preferably capable of inducing ADCC via the participation of NK cells. Thus, preferred constructs of the present invention comprising both an IgG Fc region and an IgA Fc region are preferably capable of inducing ADCC via the participation of both neutrophils and NK cells.

[0099] A significant level of ADCC activity is maintained or retained or present in the construct of the present invention. For example, a functionally effective level of ADCC is desired, e.g., at a level that allows for the killing of appropriate target cells. The ADCC induction ability of the protein construct of the present invention may be evaluated by any suitable method or assay, and the skilled person is familiar with suitable assays. The above discussion may relate to ADCC determined (or evaluated) by any suitable assay. Typically and preferably, an assay is used that can measure ADCC-induced lysis of target cells (i.e., cells expressing a target antigen recognized by the construct of the present invention are used), for example, a chromium release assay is used. Typically and preferably, ADCC activity may be expressed as the amount of specific lysis or specific cell lysis (e.g., specific lysis % or specific cell lysis %).

[0100] Another important effector function of some antibodies is the ability to induce ADCP. The ability of an antibody or other suitable protein construct to induce ADCP is generally mediated by the binding of the Fc region of the antibody or construct to the appropriate FcR on an effector cell, which is then capable of phagocytosis and thus destroying the target cell to which the antibody (or construct) is bound. In general, the ability to induce ADCP and the efficacy of ADCP induction vary depending on the Fc region of interest and whether (or to what extent) the Fc region can bind to the FcR on an effector cell capable of phagocytosis. For example, an IgG antibody can induce ADCP by binding to FcγR on an effector cell such as a macrophage. Wild-type human IgG1 Fc region and wild-type human IgG2 region generally have excellent (high) ADCP induction ability. On the other hand, IgA antibodies (eg, IgA1 and IgA2 antibodies) have only a moderate ability to induce ADCP via binding to FcαR (FcαRI) on appropriate effector cells such as macrophages.

[0101] Thus, in a preferred embodiment, the protein construct of the present invention has or retains the ability to induce antibody-dependent cellular phagocytosis (ADCP). Thus, in a preferred embodiment, the ADCP induction ability of the protein construct of the present invention is maintained or not significantly changed or not significantly decreased (e.g., compared to a suitable Fc-containing control protein, e.g., full-length IgA or full-length IgG (e.g., wild-type full-length IgA antibody or wild-type full-length IgA antibody) of a subtype selected according to the subtype of the IgA Fc region and / or IgG Fc region contained in the construct of the present invention), or the ADCP induction ability is increased (preferably significantly increased) compared to the ADCP induction ability of a suitable Fc-containing control protein (e.g., full-length IgA or full-length IgG, e.g., wild-type full-length IgA antibody or wild-type full-length IgG antibody of a suitable subtype selected according to the subtype of the IgA Fc region and / or IgG Fc region contained in the construct of the present invention).

[0102] In the construct of the present invention, a significant level of ADCP activity is maintained or retained or present. For example, a functionally effective level of ADCP, e.g., a level of ADCP that allows phagocytosis of appropriate target cells, is desired. The ADCP-inducing ability of the protein construct of the present invention may be evaluated by any suitable method or assay, and the skilled artisan will be familiar with suitable assays. The above discussion may relate to the ADCP determined (or evaluated) by any suitable assay. Typically and preferably, an assay is used that can measure ADCP-induced lysis of target cells (i.e., cells expressing a target antigen recognized by the construct of the present invention are used), for example, an assay is used that induces phagocytosis of target cells using macrophages.

[0103] Preferably, a protein construct of the invention comprising an IgA Fc region joined to an IgG Fc region has one or more, preferably all, of the functional properties described herein.

[0104] Preferably, the above abilities and properties are observed at a measurable or significant level, more preferably at a statistically significant level, when compared to an appropriate control (e.g., a control protein). In any statistical analysis described herein, it is preferred that a statistically significant difference from a relevant control or other comparable entity or measurement has a probability value of less than 0.1, preferably less than (or equal to or less than) 0.05. Methods for determining appropriate statistical significance are well known and documented in the art, and any of these may be used.

[0105] In the protein construct of the present invention, the C-terminus of the IgA Fc region is linked to the N-terminus of the IgG Fc region. Such linkage can be achieved by any convenient means, for example, directly without an intermediate or indirectly through an intermediate, such as a linker or a protein or polypeptide sequence, unit or domain, such as a structural sequence, unit or domain, such as the CH1 domain, as described elsewhere herein. Thus, the IgA Fc region and the IgG Fc region can be joined or linked to each other by any suitable method that allows each Fc region to still perform its function. In some embodiments of the present invention, the protein construct will include a linker (physical linker or linker molecule) between different parts of the construct, for example, to link the IgA Fc region to the IgG Fc region and / or to link the IgA Fc region to other parts of the construct, such as a targeting domain or an antigen binding domain, as described elsewhere herein.

[0106] Thus, in a preferred construct of the present invention, the C-terminus of the IgA Fc region is linked to the N-terminus of the IgG Fc region by a linker. Any suitable linker molecule can be used, which will be well known to those skilled in the art. For example, a peptide (or polypeptide) linker or a chemical linker or other covalent linker can be used as appropriate.

[0107] Peptide or protein (polypeptide) linkers are generally preferred. Such peptide linkers, which may contain non-natural or natural amino acids or may contain native or non-native (e.g. synthetic) sequences, are well known in the art, and therefore those skilled in the art can easily select suitable linkers with appropriate sequence, length and / or flexibility / rigidity to stably link the various components of the protein construct of the present invention together, but with correct spatial orientation or spatial optimization, so that when the individual components are joined or linked together, they retain the required functional properties mentioned above (i.e. the functional properties of each component, for example the IgA Fc region and IgG Fc region, or the function of the targeting domain or antigen binding domain as described elsewhere herein).

[0108] Thus, the linker or spacer can aid in the folding of the linked protein, and the length and / or flexibility / rigidity of the spacer or linker can be adjusted accordingly to allow for optimal or sufficient functional folding of each component. The appropriate length can be readily determined by one of skill in the art, and can be any suitable number of amino acids. However, exemplary lengths include at least 5, 10, 15, 20, 25, 30, 35, 40, or 45 amino acids long (e.g., at least 6, 7, 8, or 9 amino acids long, or at least 11, 12, 13, or 14 amino acids long), or from 5 or 10 amino acids to 50 or 60 or 70 amino acids, e.g., from 5 or 10 amino acids to 15, 20, 25, 30, 35, 40, Examples of linkers include 45, 50, 60, or 70 amino acids, or 15 to 20, 25, 30, 35, 40, 45, 50, 60, or 70 amino acids, or 20 to 25, 30, 35, 40, 45, 50, 60, or 70 amino acids, or 25 to 30, 35, 40, 45, 50, 60, or 70 amino acids, or 30 to 35, 40, 45, 50, 60, or 70 amino acids. Preferred linkers can be 15 to 30 amino acids long, for example, up to 15, 20, 25, or 30 amino acids long (or up to 40 or 50 or 60 or 70 amino acids long).

[0109] Some of the constructs of the invention are described herein as extended constructs because the tandem IgA Fc and IgG Fc regions in these constructs are further apart than those in the non-extended constructs of the invention. As described elsewhere herein, in a particular embodiment, this is achieved by including a CH1 domain (or alternative structural protein sequence, unit, or domain) and, optionally, one or more additional linkers, such as, for example, an antibody hinge region, as described herein. However, in other embodiments, this can be achieved by including a longer (extended) linker or spacer than those described above. Thus, in such embodiments, the linkers present between (or joining) the tandem IgA Fc and IgG Fc regions in the constructs of the invention can be, be at least or be up to 80, 90, 100, 105, 110, 115, 120, 125, 128, 129, 130, 135, 140, 145 or 150 amino acids in length, for example 90, 100 or 120-150 amino acids in length.

[0110] As outlined above, suitable linkers may comprise (or consist of) native or non-native (e.g., synthetic) sequences. Exemplary non-native or synthetic linkers have been reported in the art and may include linkers that comprise (or consist of) glycine and / or serine residues, such as GS linkers, which may comprise one or more repeats of a GS sequence, such as one or more repeats of a G4S linker (GGGGS (SEQ ID NO: 44)).

[0111] Exemplary native or naturally occurring peptide or polypeptide sequences that can be used as linkers will also be known to those skilled in the art. For example, it is convenient and preferred that the peptide (or polypeptide) linker used in the protein construct of the present invention to link the C-terminus of the IgA Fc region to the N-terminus of the human IgG Fc region comprises an antibody hinge region. The sequences of such hinge regions are well known and documented in the art, and exemplary sequences are shown in Table 1. For example, suitable hinge regions comprise an IgA antibody hinge region or an IgG antibody hinge region. As described elsewhere herein, it is particularly convenient to use an IgG hinge region, for example an IgG hinge region from the IgG subtype (IgG1, IgG2, IgG3, or IgG4) used in the IgG Fc region of the construct, to link the C-terminus of the IgA Fc region to the N-terminus of the IgG Fc in the construct of the present invention. Such antibody hinge regions may be used in whole or in part (e.g., fragments) and are included in the construct, provided that the functional properties of the components of the construct are retained, preferably the functional properties of the hinge region.

[0112] It would be a standard and routine procedure for one of skill in the art to select the nature of the linker and other characteristics, such as an appropriate linker length, to achieve the same (or similar) effect as observed for the linkers, e.g., hinge regions, used in the exemplified constructs. Exemplary preferred lengths are provided elsewhere herein. In other preferred embodiments of the invention, particularly in the extended constructs of the invention, the C-terminus of the IgA Fc region can be linked to the N-terminus of the IgG Fc region by a CH1 domain. In other words, the CH1 domain can be used as an intermediate to link the IgA Fc region and the IgG Fc region. In such embodiments, any CH1 domain can be used, some examples of which are shown in Table 1, such as the CH1 domain of IgG or IgA. In a preferred embodiment, the CH1 domain is an IgG CH1 domain, such as an IgG1 CH1 domain. Such preferred intermediates may include a CH1 domain and may further include other linker or spacer elements, such as those described above and elsewhere herein. By way of example, one or more linkers, such as peptide linkers, may be present, such as one or more antibody hinge regions. In such an embodiment, the CH1 domain is located between the IgA Fc region and the IgG Fc region in the tandem portion of the construct, and a linker can be used to conveniently link the N-terminus of the CH1 domain to the C-terminus of the IgA Fc region and / or the C-terminus of the CH1 domain to the N-terminus of the IgG Fc region. As described elsewhere herein, such a linker can be an antibody hinge region. As in other embodiments, it is particularly convenient to use a hinge region that matches the nature of the CH1 domain. Thus, if the CH1 domain is an IgG CH1 domain, in some embodiments, the linker will be an IgG hinge region from the IgG subtype (IgG1, IgG2, IgG3, or IgG4) used in the IgG CH1 domain. In a preferred embodiment, the CH1 domain is an IgG1 CH1 domain. In such and other embodiments, the preferred linker is an IgG1 hinge region.A preferred such extended construct is shown in Example 2, which comprises, for example, from the N-terminus to the C-terminus, an IgA Fc region (e.g., IgA2), an IgG hinge region (e.g., an IgG1 hinge region), a CH1 domain (e.g., an IgG1 CH1 domain), a further IgG hinge region (e.g., an IgG1 hinge region), and an IgG Fc region (e.g., an IgG1 Fc region).

[0113] In these and other embodiments of extended constructs of the invention, the preferred overall length of the region of the construct joining (between) the C-terminus of the IgA Fc region and the N-terminus of the IgG Fc region is as described elsewhere herein for linkers or spacers used in the constructs of the invention, e.g., at least or up to 150 amino acids in length.

[0114] An advantage of using a peptide (or polypeptide) linker is that it allows the multiple polypeptide chains in the construct to be produced as a single polypeptide, for example as a fusion protein or polypeptide.

[0115] The terms "fusion protein", "fusion polypeptide" and the like are used herein to denote the functional joining of two or more protein components in the same polypeptide sequence or in the same open reading frame (ORF). Some examples of such fusion proteins can also be described as gene fusions, since they are encoded by the same nucleic acid sequence (sometimes called "fusion genes" or "fusion nucleotide sequences"). In such fusion proteins, the two (or more) protein components (or the nucleic acid sequences encoding them) can be directly adjacent, but also and preferably, the components can be joined by a spacer or linker consisting of a suitable peptide or polypeptide, e.g., as described above.

[0116] Since the constructs of the invention comprise two Fc dimers in tandem or adjacent to each other, in such embodiments the protein construct conveniently comprises two fusion polypeptides (or fusion polypeptide chains), in such embodiments each fusion polypeptide comprises (i) one chain of an IgA Fc region (i.e. one chain of an IgA Fc dimer) and (ii) one chain of an IgG Fc region (i.e. one chain of an IgG Fc dimer), said polypeptide chain of the IgG Fc region being arranged (or located) C-terminally to said polypeptide chain of the IgA Fc region.

[0117] Thus, in such an embodiment, the polypeptide chain (fusion polypeptide) typically comprises, from the N-terminal end to the C-terminal end, a polypeptide comprising half of an IgA Fc dimer and a polypeptide comprising half of an IgG Fc dimer. As described elsewhere herein, the N-terminal end of the polypeptide constituting the IgA Fc region (or dimer) may be linked (or fused) to other entities, such as suitable targeting domains or antigen-binding domains. Thus, the protein constructs of the invention (or the IgA Fc region and IgG Fc region components of the constructs) can be used to create Fc fusions or Fc fusion proteins with other suitable entities. In other words, the IgA Fc region and IgG Fc region components of the constructs linked together can be, for example, a polypeptide unit that can be fused to any polypeptide or protein of interest.

[0118] The skilled artisan is familiar with methods for producing fusion polypeptides, including, for example, expressing (e.g., in a host cell) a nucleic acid molecule encoding the fusion polypeptide. Such a nucleic acid molecule typically comprises a contiguous nucleotide sequence encoding in frame the various components of the fusion polypeptide. In the present case, such a nucleic acid molecule may comprise a contiguous nucleotide sequence comprising a nucleotide sequence encoding a polypeptide constituting one chain of an IgA Fc region and a nucleotide sequence encoding a polypeptide constituting one chain of an IgG Fc region, said nucleotide sequence encoding said chain of the IgG Fc region being located at the 3' end of the nucleic acid molecule encoding said chain of the IgA Fc region. Then, when expressed, dimerization can occur between the two polypeptide chains to form a complete IgA region and an IgG Fc region (dimer).

[0119] While the above description focuses on the presence of a linker or spacer between the IgA Fc region and the IgG Fc region of the construct, a linker sequence may be suitably included elsewhere in the construct of the invention, for example, between other components that may be present in the construct. Thus, in some constructs of the invention, a linker may be included between the N-terminus or N-terminal end of the IgA Fc region and any other desired entity present at that location, such as a targeting domain or an antigen-binding domain as described elsewhere herein. Such linkers may be in the form as described above, for example, a native or non-native (synthetic) peptide linker. A convenient and preferred peptide (polypeptide) linker for this connection may be in the form of an antibody hinge region, conveniently an IgA antibody hinge region, for example an IgA hinge region from an IgA subtype (IgA1, IgA2) used in the IgA Fc region of the construct. However, as described elsewhere herein, in other embodiments, an IgG hinge region, for example an IgG1 hinge region, may be used. All or parts (eg fragments) of such antibody hinge regions may be used, provided that the functional properties of the components of the construct are retained, preferably the functional properties of the hinge region.

[0120] Although peptide (polypeptide) linkers are convenient and preferred in some embodiments, any other suitable means of linking or bonding may be used, including, for example, any other form of linker, including chemical linkers, e.g., chemical cross-linkers, provided that the functional properties (described elsewhere herein) of the various components to be linked are preserved when the individual components are linked or joined together. Suitable cross-linking agents and methods for linking (or binding or joining or linking or conjugating) different proteins or polypeptides together are known in the art.

[0121] Thus, in some embodiments, the IgA Fc region and the IgG Fc region of the constructs of the invention may be produced separately and then linked (or joined or joined or linked or conjugated) to one another.

[0122] In some embodiments, there is no linker. Thus, for example, in some embodiments, a polypeptide comprising an IgA Fc region may be directly joined to a polypeptide comprising an IgG Fc region in accordance with the present invention. Such direct joining, in terms of the fusion polypeptide, may result in the first (i.e., N-terminal) amino acid of the IgG Fc region being fused directly (by a peptide bond) to the last (i.e., C-terminal) amino acid of the IgA Fc region (i.e., there is no linker in between).

[0123] Any IgG Fc region or IgA Fc region can be used in the protein construct of the present invention. Thus, in some embodiments, these regions will comprise or correspond to wild-type or native sequences. However, in other embodiments, the IgG Fc region and / or IgA Fc region can comprise mutations or modifications, in particular mutations or modifications that increase or enhance the effector function of the Fc region or mutations or modifications that increase or enhance the plasma half-life. Suitable mutant or modified Fc regions are well known and reported in the art, and any of these may be used.

[0124] Preferred mutations or modifications include those that increase or enhance binding to an Fc receptor, preferably those that increase or enhance binding to FcRn, or those that increase or enhance C1q binding.

[0125] In some embodiments, a preferred Fc region for use in the constructs of the invention (an IgG Fc region, such as an IgG1 Fc region, an IgG2 Fc region, an IgG3 Fc region, or an IgG4 Fc region) is a modified IgG Fc region characterized by comprising one or more of the following modifications: (i) an arginine (R) residue or a similar residue (such as a lysine (K) residue) at position 311 (or a position corresponding thereto); (ii) a glutamic acid (E) residue or a similar residue (such as an aspartic acid (D) residue) at position 428 (or a position corresponding thereto); and (iii) at position 434 (or a position corresponding thereto), a tryptophan (W) residue or a similar residue, such as a tyrosine (Y) residue or a phenylalanine (F) residue.

[0126] These positions are defined using standard EU numbering of IgG regions, with ASTK in the CH1 domain starting at position 118 (see, for example, http: / / www.imgt.org / IMGTScientificChart / Numbering / Hu_IGHGnber.html#refs). Thus, the positions of these residues (or their corresponding positions) in the various constructs of the invention can be easily determined. Indeed, these modified IgG Fc regions are also referred to herein as REW mutations, and some exemplary sequences are shown in Table 1. The wild-type residues at positions 311, 428, and 434 are Q, M, and N for IgG1 and IgG2, respectively.

[0127] In preferred embodiments of the invention, two or more of the above modifications are present, most preferably all three modifications are present. Such modifications are further described in WO 2017 / 158426, but are examples of mutations or modifications that have the effect of increasing or enhancing pH-dependent binding to FcRn. The inventors have also shown herein that such modifications can increase binding to C1q. They have also shown increased CDC induction. In some embodiments, such modifications are preferably used in IgG2 Fc regions. In some embodiments, such modifications are preferably used in IgG1 Fc regions. In some embodiments, such modifications are preferably used in IgG3 Fc regions, optionally with the R435H mutation described below. In some embodiments, such modifications are preferably used in IgG4 Fc regions. In some embodiments, REW substitutions are preferred.

[0128] Another preferred possible additional modification that can be used in the IgG Fc region in the constructs of the invention includes the R435H mutation as described in Stapleton et al., 2011, Nat. Comm. 20(2):599. Such a modification also has the effect of increasing binding to FcRn and is preferred for use when the construct comprises an IgG3 Fc region. Again, this position is defined using the EU numbering system as described above.

[0129] In a preferred embodiment, the construct of the present invention further comprises a targeting domain (or targeting unit), for example, further binds to a targeting domain (or targeting unit). Such a targeting domain can be any entity that can bind, for example, specifically bind, to a desired target molecule or target entity. Thus, such a targeting domain can be a receptor, preferably a receptor domain (for example, a receptor or receptor domain for a ligand) or a receptor ligand (for example, a ligand for a receptor or receptor ligand), or an antigen binding domain. Preferred targeting domains and target molecules are proteins or polypeptides. Thus, preferred targeting domains are binding proteins.

[0130] In some embodiments, the targeting domain of the protein construct of the invention is or comprises an antigen-binding domain.

[0131] Preferred antigen-binding domains for use in the protein constructs of the present invention are antibodies or antibody fragments, including antigen-binding fragments of antibodies.

[0132] Thus, in a preferred embodiment of the invention, the protein construct comprises an antibody (or immunoglobulin), or an antigen-binding fragment thereof.

[0133] Since the N-terminal region of the protein construct of the present invention comprises an IgA Fc region, a particularly preferred antibody is an IgA antibody or an antigen-binding fragment thereof, in which case the Fc region is already provided in the construct. In some embodiments, the IgA antibody is an IgA1 antibody. In some embodiments, the IgA antibody is an IgA2 antibody. In other embodiments, an IgG antibody or an antigen-binding fragment thereof is used as the antigen-binding domain. In some embodiments, the IgG antibody is an IgG1 antibody. In some embodiments, the IgG antibody is an IgG2 antibody. In some embodiments, the IgG antibody is an IgG3 antibody. In some embodiments, the IgG antibody is an IgG4 antibody. The subunit structures and three-dimensional configurations of different classes of antibodies are well known.

[0134] As used herein, the terms "antibody" and "immunoglobulin" refer broadly to any immunological binding agent that contains an antigen-binding domain. The terms include antibody fragments that contain an antigen-binding domain.

[0135] As will be appreciated by those of skill in the art, immunological binding reagents encompassed by the term "antibody" include or are intended to cover all antibodies and antigen-binding fragments thereof, including whole antibodies (i.e., full-length antibodies), dimeric, trimeric, and multimeric antibodies; bispecific antibodies; chimeric antibodies; recombinant and engineered antibodies, and fragments thereof.

[0136] Techniques for preparing and using various antibody-based constructs and fragments are well known in the art. Monoclonal antibodies are particularly preferred.

[0137] The antibody or antibody fragment comprises an antibody light chain variable region (V L ) and an antibody heavy chain variable region (V H) are preferably included. In summary, although the six CDRs generally confer antigen-binding specificity to the antibody, single domain antibodies having only three CDRs can be used as well, and even one or two CDR regions can mediate antigen binding, especially when only low or moderate affinity binding is desired. The heavy and light chain variable regions also have four framework regions (FR1, FR2, FR3 and FR4 from the amino to carboxy termini). These framework regions space the CDRs apart. Although the VL and VH generally form an antigen-binding site, it is also possible to use single domain antibodies that generally only contain three CDR domains, e.g., a VL region or a VH region capable of binding to an antigen.

[0138] The CDRs of the antibody used in the construct of the present invention are preferably separated from each other by suitable framework regions as found in naturally occurring antibodies and / or effectively engineered antibodies. Thus, the CDR sequences are preferably provided or incorporated into a suitable framework or scaffold to allow antigen binding. As such framework sequences or framework regions, the naturally occurring framework regions FR1, FR2, FR3, and / or FR4 may be used as appropriate to form a suitable scaffold, or a consensus framework region, for example, identified by comparing various naturally occurring framework regions, may be used. Alternatively, non-antibody scaffolds or frameworks may be used, such as T cell receptor frameworks.

[0139] The antibody or antibody fragment for use in the construct of the invention may further comprise an antibody constant region, e.g., a heavy chain constant region and / or a light chain constant region, in addition to the appropriate variable domains, including the CDR and FR sequences constituting the antigen-binding site. The heavy chain constant region comprises the three constant domains CH1, CH2, and CH3, and the light chain constant region comprises the CL constant domain. In some embodiments of the invention, the antibody (or fragment) will further comprise a CH1 domain and / or a CL domain. In a preferred embodiment, these domains will be derived from an IgA antibody. Thus, a preferred construct of the invention comprises an IgA Fc region, a VL region and / or a VH region, as well as a CH1 region and a CL region, preferably a CH1 region and a CL region corresponding to or derived from an IgA antibody. In another preferred embodiment, these domains will be derived from an IgG antibody. Thus, a preferred construct of the invention comprises an IgA Fc region, a VL region and / or a VH region, as well as a CH1 region and a CL region, preferably a CH1 region and a CL region corresponding to or derived from an IgG antibody.

[0140] Suitable sequences for such constant regions are well known and documented in the art.

[0141] When a full complement of constant regions from heavy and light chains are included in a construct of the invention, such a construct is typically referred to as comprising a "full-length" or "whole" antibody. In a preferred embodiment, the protein construct comprises a whole (or full-length) antibody, preferably a whole (or full-length) IgA antibody. A "whole" or "full-length" antibody comprising two heavy chains and two light chains is preferred in some embodiments. When a full-length IgA antibody is included, the IgA Fc region of the full-length IgA antibody can provide the IgA Fc region of the claimed construct.

[0142] However, in some embodiments, the antigen-binding domain of the protein construct does not comprise a whole or full-length antibody. In some embodiments, the protein construct comprises an antigen-binding fragment of an antibody, preferably a fragment of an IgA antibody. In other preferred embodiments, a fragment of an IgG antibody can be used. Such an antigen-binding fragment may comprise three or six CDRs, e.g., an sdAb antibody or an scFv antibody or an Fv antibody, and may also comprise a CH1 region and / or a CL region, e.g., a Fab fragment. Suitable antigen-binding fragments and formats of antibodies are known in the art, any of which may be used. Advantageously, the protein construct of the invention comprises two antigen-binding domains (or other targeting domains), one each tethered or attached to each chain of the IgA Fc region / IgA Fc fragment, so that, for example, the antigen-binding domain can be in the form of a Fab2 fragment. Indeed, in some embodiments, e.g. in the extended constructs of the invention as described elsewhere herein, the use of a Fab2 fragment of an IgG, e.g. a Fab2 fragment of an IgG1, is preferred, e.g. in combination with the presence of a CH1 domain (or alternative structural protein sequence, unit or domain) or an extended linker sequence as described elsewhere herein, as an intermediate linking the IgA Fc region and the IgG Fc region of the construct. These two antigen binding domains (or other targeting domains) may be identical or different. Similarly, it is possible to use only one antigen binding domain (or other targeting domain), e.g., to be linked to only one chain of the IgA Fc region / fragment.

[0143] The protein construct may be, for example, an sdAb (e.g., a nanobody antibody or H In some embodiments comprising an antigen-binding fragment or targeting domain composed of a single polypeptide chain, such as an H antibody, or a VH antibody or a VL antibody) or an scFv fragment, the protein construct of the invention may comprise (or consist of) two polypeptide chains.

[0144] In some embodiments where the protein construct comprises an antigen-binding fragment or targeting domain made up of two polypeptide chains, e.g., a Fab fragment, or, for example, provided by a full-length antibody, the protein construct of the invention can comprise (or consist of) four polypeptide chains.

[0145] Thus, according to the present invention, a protein construct comprises (or consists of) two polypeptide chains constituting (or forming) an IgA Fc region and an IgG Fc region, e.g. constituting (or forming) adjacent or fused tandem Fc regions present in a construct of the invention.

[0146] In some embodiments, an antibody (or antigen-binding fragment) for use in a construct of the invention has been reformatted (e.g., from an IgG antibody or other class of antibody) into an IgA format (e.g., an IgA1 or IgA2 format). Thus, in some embodiments, the antibody (or antigen-binding fragment) is an IgA antibody that comprises an antigen-binding domain of a non-IgA antibody (e.g., obtained from, derived from, or based on an antigen-binding domain of a non-IgA antibody, e.g., an antigen-binding domain of an IgG antibody). Methods for reformatting antibodies into an IgA format are well known in the art and one of ordinary skill in the art would be familiar with such methods.

[0147] In some embodiments, IgA antibodies (or antigen-binding fragments thereof) for use in the constructs of the invention have been modified (or mutated) to maintain a monomeric form (e.g., to prevent dimerization). Suitable modifications in this regard include IgA antibodies with modified (or mutated or inactivated or truncated) tails, or tail-free (or tail-removed) IgA antibodies, as described elsewhere herein.

[0148] In preferred embodiments, the sequences constituting the targeting domains, e.g., receptors, receptor domains, ligands, receptor ligands, antigen-binding domains, antibodies (or antigen-binding fragments thereof) for use in the constructs of the invention are human sequences. In this regard, human sequences, e.g., human antibodies, generally have potential advantages for use in human therapies, e.g., the human immune system should not recognize the antibodies as foreign.

[0149] In other preferred embodiments, as set forth elsewhere herein, the preferred IgA Fc region and / or IgG Fc region is a human Fc region.

[0150] The term "human" as used herein in reference to antibody molecules refers to the human H area, V L "human" refers to an antibody having a constant antibody region (CDR region, CDR region, or FR region) and preferably a constant antibody region, isolated from or derived from a human repertoire, or derived from or corresponding to a sequence found in a human or human repertoire, e.g., in a human germ cell or human somatic cell, or in a body fluid, or in a human antibody library, such as a phage display library. Similarly, the term "human" as used herein in reference to a targeting domain (or binding protein) or Fc region refers to a protein sequence isolated from or derived from a human, or corresponding to a sequence found in a human, e.g., in a human germ cell or human somatic cell. Thus, such human sequences can be obtained from a human sample, e.g., a human body fluid or a human cell. Humanized sequences, e.g., humanized targeting domains, antigen binding domains, antibodies or antibody fragments, or humanized Fc regions, can also be used in the constructs.

[0151] Non-antibody or non-immunoglobulin based targeting domains or binding proteins can also be used in the constructs of the invention and can themselves be selected for their ability to specifically bind to a particular target molecule or antigen. Such molecules are also referred to as antibody mimics (or antibody mimetics). Examples of suitable non-immunoglobulin-based targeting domains or binding proteins are known and reported in the art and include: fibronectin (or fibronectin-based molecules), such as Adnectin (e.g., Compound Therapeutics, Inc., Waltham, Massachusetts), e.g., based on the 10th module of the fibronectin type III domain; affimers (e.g., Avacta); ankyrin repeat proteins (e.g., Molecular Partners AG, Zurich, Switzerland); lipocalins, e.g., anticalin (e.g., Pieris Proteolab AG, Freising, Germany); human A domains (e.g., Avimers); Staphylococcal protein A (e.g., Affibody AG, Sweden); thioredoxin; and gamma-B-crystallin or ubiquitin-based molecules, such as affilin (e.g., Scil Proteins GmbH, Halle, Germany). As mentioned above, such molecules can also be used as scaffolds onto which suitable CDRs that mediate target antigen binding can be grafted. For example, the CDRs of a suitable immunoglobulin-based targeting domain or binding protein can be grafted onto a suitable non-immunoglobulin scaffold.

[0152] Typically, the targeting domain is located at the N-terminal end or N-terminal region of the construct of the invention. Thus, in a preferred embodiment, the targeting domain is attached to the N-terminus of the IgA Fc region of the construct, for example, to the N-terminus or N-terminal end of the IgA Fc region. Such attachment can be direct or indirect, for example, via a linker, preferably via an antibody hinge region, as described elsewhere herein. Of course, it is important that the ability of the targeting domain to bind to its target is retained (or not significantly affected) when incorporated into the construct of the invention.

[0153] Preferred target molecules to which the targeting domain binds are therapeutically (or clinically) relevant target proteins (or target antigens). Thus, in some embodiments, the protein construct of the present invention binds to a disease-associated target protein (or target antigen) by the targeting domain. The disease-associated target protein (or target antigen) may be a target protein (or antigen) whose expression (e.g., undesired expression or aberrant expression or overexpression) is associated with a disease. In some embodiments, the disease is a cancer (or tumor), e.g., a solid tumor such as breast cancer, or a blood cancer. Alternatively, the disease is caused by a pathogen, e.g., an infectious pathogen such as a bacterium.

[0154] Thus, in some embodiments, the targeting domain binds to a given cancer (or tumor) protein or antigen (e.g., a cancer-specific protein or antigen, or a cancer-associated protein or antigen, or a tumor-specific protein or antigen, or a tumor-associated protein or antigen). In some embodiments, the cancer is breast cancer. In other embodiments, the targeting domain binds to a protein or antigen associated with an infectious agent, e.g., a bacterium.

[0155] Thus, in some embodiments, the targeting domain, e.g., the antigen binding domain, binds to a target molecule on a cancer cell or an infectious pathogen. Preferably, the cancer cell is from a solid tumor or a blood cancer, or the infectious pathogen is a bacteria. In some embodiments, the cancer or solid tumor is a breast cancer.

[0156] Exemplary cancer-related targets are members of the human epidermal growth factor receptor.Thus, in some embodiments, the construct of the present invention, for example, the construct that comprises a targeting domain (e.g., an antigen-binding domain), binds to HER2.Overexpression of HER2 protein may play an important role in the development and progression of certain types of breast cancer.

[0157] In some embodiments, the construct of the invention comprises the antigen-binding domain of an anti-HER2 antibody, for example the antibody trastuzumab.

[0158] Yet another exemplary cancer-associated target, for example, associated with cancerous B-cell lines and thus suitable for the treatment of hematological cancers, includes CD20.Thus, in some embodiments, the construct of the present invention, for example, a construct comprising a targeting domain (e.g., an antigen-binding domain), binds to CD20.Thus, a preferred construct comprises an anti-CD20 antibody or an antigen-binding fragment thereof.Overexpression of CD20 protein may play an important role in the development and progression of certain hematological cancers, particularly those involving B cells.

[0159] In another aspect, the invention provides a method for increasing the in vivo half-life (e.g. plasma half-life or serum half-life) of a protein (preferably an IgA antibody or fragment thereof, or an alternative targeting domain as described herein), comprising incorporating the protein etc. into a protein construct of the invention (preferably at the N-terminal end of the construct, e.g. linked to the N-terminus of an IgA Fc region) or tethering a protein construct of the invention (or a polypeptide unit as described above) to said protein etc. (preferably at the C-terminal end of said protein). The statements made with respect to various features of the proteins and protein constructs of the other aspects and preferred embodiments of the invention also apply mutatis mutandis to this aspect of the invention.

[0160] The Fc region used in the constructs of the invention can be obtained, derived or correspond to an Fc region from any source or species, or can be a fragment or variant thereof, provided that it retains the ability to bind to an Fc receptor or induce effector function. Mammalian sources or species are preferred, and any suitable mammalian source or species may be used, such as humans or any farm animal, domestic animal or laboratory animal. Specific examples include mice, rats, pigs, cats, dogs, sheep, rabbits, horses, cows, and non-human primates (e.g., cynomolgus monkeys). However, the mammal is preferably a human. Sequences of Fc regions from various species are known in the art, and thus suitable Fc regions for use in the present invention can be readily generated or produced by standard techniques, such as recombinant techniques. While the above description has focused on illustrating protein constructs of the present invention, said protein constructs may be conveniently prepared or manufactured using appropriate nucleic acid molecules encoding all or part of such protein constructs.

[0161] It will thus be appreciated that a nucleic acid molecule, e.g. one or more nucleic acid molecules (e.g. a set of nucleic acid molecules) comprising a nucleotide sequence encoding a protein construct of the invention, e.g. a recombinant protein construct as defined herein, or a portion thereof (e.g. a single chain or the first or second strand of the protein construct), forms a further aspect of the invention. An expression vector comprising, e.g. one or more such nucleic acid molecules, and a host cell comprising said expression vector or nucleic acid molecule or protein construct form further aspects.

[0162] Typically, the one or more nucleic acid fragments encoding a protein construct of the present invention are incorporated into one or more suitable expression vectors to facilitate production of the protein construct, e.g., a recombinant protein construct.

[0163] Thus, the present invention contemplates an expression vector, e.g., one or more expression vectors, e.g., one or more recombinant expression vectors, that contain or include one or more (or set of) nucleic acid molecules of the present invention and the necessary regulatory sequences for the transcription and translation of the protein sequence encoded by the nucleic acid molecules of the present invention. The vector may further include sequences that allow antibiotic resistance and replication of the vector. Suitable vectors and regulatory sequences will be well known to those skilled in the art.

[0164] The expression vector of the present invention, e.g., a recombinant expression vector, or the nucleic acid molecule of the present invention can be introduced into a host cell to produce a transformed host cell. The terms "transformed with", "transfected with", "transformation" and "transfection" are intended to encompass the introduction of a nucleic acid (e.g., a vector) into a cell by one of many possible techniques known in the art. Suitable methods for transforming and transfecting host cells are described in Sambrook et al., 1989 (Sambrook, Fritsch and Maniatis, Molecular Cloning: A Laboratory Manual, 2nd Ed., Cold Spring Harbor Press, Cold Spring Harbor, NY, 1989) and other laboratory texts.

[0165] Suitable host cells include a wide variety of eukaryotic host cells and prokaryotic cells. For example, the molecules of the invention can be expressed in yeast cells, or mammalian cells, or prokaryotic cells such as E. coli or Pichia pastoris.

[0166] The protein constructs and nucleic acid molecules of the present invention are generally "isolated" or "purified." The terms "isolated" or "purified" typically refer to a protein or nucleic acid that is substantially free of cellular material or other proteins (or other nucleic acids) from the source from which it is derived or produced.

[0167] The skilled artisan will appreciate that the protein constructs of the invention may be prepared by any of several methods well known and described in the art, but are most preferably prepared using recombinant methods. For example, the various components of the construct may be suitably encoded on a single polypeptide chain or on multiple polypeptide chains, which are then joined or linked, or which are otherwise associated with one another to form the protein constructs of the invention.

[0168] Thus, yet another aspect of the present invention provides a method for producing a protein construct of the present invention, comprising culturing a host cell of the present invention. A preferred method comprises (i) culturing a host cell comprising one or more expression vectors of the present invention or one or more nucleic acid sequences of the present invention under conditions suitable for the expression of the encoded protein construct, and (ii) an optional step of isolating or obtaining the expressed protein construct from said host cell or growth medium / supernatant. Such a production method may further comprise a step of purifying the protein product and / or formulating the protein product into a composition comprising at least one additional component, such as a pharma- ceutically acceptable carrier or excipient.

[0169] Preferred protein constructs of the invention generally comprise two identical polypeptide chains (e.g., where the targeting domain is absent or is a single polypeptide chain, such as a single chain antibody) or a pair of identical polypeptide chains (e.g., where the targeting domain is composed of two polypeptide chains, such as an antibody or antibody fragment having separate heavy and light chains), and thus in such embodiments, a single or two polypeptide chains are expressed, as appropriate, in a host cell, and the complete protein construct of the invention can be assembled in the host cell and isolated or purified therefrom. Protein constructs of the invention can be produced, purified or isolated by standard methods well known to those of skill in the art.

[0170] The present invention also provides various conjugate proteins and fragments thereof that are functionally linked to at least one other agent (e.g., therapeutic agent) that is different from any targeting domain present in the construct of the present invention. The term "immunoconjugate" is used broadly to define the functionally effective association of a protein construct with another effective agent (e.g., therapeutic agent). Recombinant fusion proteins are particularly contemplated. Any mode of linking may be suitable, as long as the targeting domain can bind to the target and the other agent functions sufficiently upon delivery.

[0171] In some embodiments, the protein construct of the invention is used (e.g., used therapeutically) in its "naked" form, not immunoconjugated, e.g., comprising a targeting domain together with tandem IgA Fc regions and IgG Fc regions.

[0172] A composition comprising a protein construct of the invention (or one or more nucleic acid molecules of the invention, or one or more expression vectors of the invention, or one or more immunoconjugates of the invention) constitutes yet another aspect of the invention. A formulation (composition) comprising one or more protein constructs of the invention (or one or more nucleic acid molecules of the invention, or one or more expression vectors of the invention, or one or more immunoconjugates of the invention) in admixture with a suitable diluent, carrier or excipient constitutes a preferred embodiment of the invention. Such a formulation may be for pharmaceutical use (is a pharmaceutical composition), and therefore the composition of the invention is preferably a pharma-ceutical acceptable composition. Suitable diluents, excipients and carriers are known to those skilled in the art.

[0173] The compositions of the present invention may be in a form suitable for, for example, oral, nasal, parenteral, intravenous, topical, or rectal administration. Unless otherwise specified, administration is typically by parenteral route, preferably by subcutaneous, intramuscular, intracapsular, intrathecal, intraperitoneal, intratumoral, transdermal, or intravenous injection. In some embodiments, subcutaneous administration is preferred.

[0174] The compositions of the invention as defined herein may be in conventional pharmaceutical dosage forms such as coated tablets, nasal or pulmonary sprays, solutions, liposomes, powders, capsules, or sustained release forms, etc. Conventional pharmaceutical excipients, as well as conventional manufacturing methods, may be used to prepare these forms.

[0175] The injection solutions may be prepared in a conventional manner, e.g., by adding suitable preservatives or stabilizers, and then filled into vials or ampoules for injection.

[0176] Nasal drops may similarly be formulated as an aqueous solution and packaged in a spray container with an aerosol propellant or packaged in a spray container equipped with a manual compression means.

[0177] Parenteral administration may be performed by subcutaneous, intramuscular or intravenous injection using a syringe, optionally a pen-type syringe. Alternatively, parenteral administration may be performed by an infusion pump. Another option is a composition, which may be a powder or a liquid, for administering the molecule or protein construct in the form of a nasal or pulmonary spray. As a further option, the molecule or protein construct of the invention may be administered transdermally, for example from a patch, optionally from an iontophoretic patch, or transmucosally, for example bucally.

[0178] The appropriate dosage unit can be determined by one of skill in the art.

[0179] The pharmaceutical compositions may, for simultaneous administration or combination regimens, additionally comprise other active ingredients, such as those described elsewhere herein.

[0180] The protein constructs of the invention as defined herein may be used as molecular tools for in vitro or in vivo applications and assays.Thus, a further aspect of the invention provides reagents comprising the protein constructs of the invention (or other molecules) as defined herein, and the use of such protein constructs (or other molecules) as molecular tools, for example in in vitro or in vivo assays.

[0181] A further aspect of the invention provides a protein construct of the invention (e.g. a construct comprising a targeting domain, e.g. an IgA antibody or fragment thereof) for use in therapy. Therapy includes treatment (e.g. treatment of an existing disease) or prevention (prophylaxis). In some embodiments, active treatment of an existing disease is preferred.

[0182] In some embodiments, the present invention provides a protein construct of the invention (e.g. a construct comprising an IgA antibody or an IgG antibody or a fragment thereof) comprising a targeting domain that binds to a predetermined target or antigen, e.g. a receptor, receptor domain, ligand, receptor ligand, or antigen binding domain, for use in the treatment of a disease characterized by (or associated with) expression of said target or antigen on the cell surface (e.g. unwanted or aberrant expression of said target or antigen).

[0183] For example, in some embodiments, the invention provides a protein construct of the invention (e.g., a construct comprising an IgA or IgG antibody or a fragment thereof) comprising a targeting domain, e.g., a receptor, ligand, or antigen-binding domain, that binds to a given cancer (or tumor) antigen (e.g., a cancer-specific or cancer-associated or tumor-specific or tumor-associated antigen), for use in treating cancer (or tumor). In some embodiments, the cancer or tumor antigen is CD20, and the targeting domain used is one that binds to CD20.

[0184] For example, in some embodiments, the present invention provides a protein construct of the present invention (e.g., a construct comprising an IgA or IgG antibody or a fragment thereof) comprising a targeting domain, e.g., a receptor, ligand, or antigen-binding domain, that binds to a given molecule or antigen (e.g., a pathogen-specific or pathogen-associated antigen) expressed by a pathogen, for use in treating an infectious disease caused by or associated with said pathogen. A preferred infectious pathogen is a bacterium.

[0185] In some embodiments, the present invention provides a protein construct of the present invention (e.g., a construct comprising an IgA or IgG antibody or a fragment thereof) comprising a targeting domain that binds to Her2 protein, such as a receptor, a receptor domain, a ligand, a receptor ligand, or an antigen binding domain, for use in treating Her2 positive cancer, such as Her2 positive breast cancer. In some embodiments, the antigen binding domain (VL domain and VH domain) is or is based on the antibody trastuzumab.

[0186] In another embodiment, the present invention provides a protein construct of the present invention for use in mucosal delivery, e.g. therapeutic uses as described herein are achieved by mucosal delivery of the protein construct of the present invention by a suitable route of administration, e.g. intranasal or intrapulmonary administration.

[0187] In another aspect, the invention provides an immunoconjugate of the invention for use in therapy, such as a therapy as described elsewhere herein.

[0188] The present invention further provides the use of a protein construct of the invention, preferably a recombinant protein construct of the invention, in the manufacture of a medicament or composition for use in therapy or for use in the treatment or prevention of any of the above mentioned diseases or conditions.

[0189] The present invention further provides a method for the treatment or prevention of any of the above-mentioned diseases or conditions, comprising the step of administering a therapeutically effective amount of a protein construct, preferably a recombinant protein construct, of the present invention to a subject in need thereof.

[0190] Alternative and preferred embodiments and features of the invention, as described elsewhere herein, particularly with respect to therapeutic uses, apply equally to these therapeutic methods and uses of the invention.

[0191] The nucleic acid molecules or expression vectors of the invention may likewise be used in the methods of treatment or uses described herein.

[0192] Treatment of a disease or condition (e.g., treatment of an existing disease) according to the present invention includes curing the disease or condition, or any alleviation or remission of the disease (e.g., reducing the severity of the disease) or any alleviation or remission of the symptoms of the disease.

[0193] As will be clear from the disclosure elsewhere herein, the methods and uses of the present invention are suitable for the prevention of disease as well as for the active treatment of disease (e.g., the treatment of an existing disease). Thus, preventive treatment is also encompassed by the present invention. Thus, in the methods and uses of the present invention, treatment also includes, where appropriate, prophylactic treatment or prevention. Such preventative (or protective) aspects may be conveniently performed on healthy or normal, or at-risk subjects, and may include both complete prevention and significant prevention. Similarly, significant prevention may include scenarios in which the severity of a disease or symptoms of a disease are reduced (e.g., measurably or significantly reduced) compared to the severity or symptoms expected if the treatment is not administered.

[0194] The protein constructs and compositions and methods and uses of the present invention may be used in combination with other therapeutic agents or therapeutic agents.

[0195] Thus, the "combination" embodiment of the invention includes, for example, the use of a protein construct of the invention in combination with a drug or therapeutic agent that is not functionally tethered to the protein construct. In other "combination" embodiments of the invention, the protein of the invention is an immunoconjugate in which the protein construct of the invention itself is functionally associated or combined with a drug or therapeutic agent to be used in combination. Functionally tethering includes all forms of direct and indirect tethering described herein and known in the art.

[0196] Thus, the present invention provides compositions, pharmaceutical compositions, therapeutic kits, and pharmaceutical cocktails comprising, optionally in at least a first composition or container, a biologically effective amount of at least a first protein construct of the present invention and a biologically effective amount of at least a second biological agent, which is often, but need not be, a therapeutic agent.

[0197] When the at least second biological agent includes a therapeutic agent, such therapeutic agent will typically be for use in connection with the treatment of one or more of the disorders defined above.

[0198] Thus, in certain embodiments, "at least a second therapeutic agent" will be included in the therapeutic kit or cocktail, the term being chosen in view of the fact that the protein construct of the invention is the first therapeutic agent.

[0199] In certain embodiments of the invention, the second therapeutic agent may be a radiotherapeutic agent, a chemotherapeutic agent, angiogenesis inhibitor, an apoptosis inducer, an antitubulin agent, an anti-cellular or cytotoxic agent, a steroid, a cytokine antagonist, a cytokine expression inhibitor, a chemokine antagonist, a chemokine expression inhibitor, an ATPase inhibitor, an anti-inflammatory agent, a signal pathway inhibitor, a checkpoint inhibitor, an anti-cancer agent, another antibody, or a coagulant.

[0200] Generally speaking, the at least second therapeutic agent may be administered to the subject substantially simultaneously with the protein construct of the present invention, for example from a single pharmaceutical composition or from two pharmaceutical compositions administered together with no separation in between.

[0201] Alternatively, the at least second therapeutic agent can be administered to the subject sequentially with the administration of the protein of the invention. As used herein, "sequentially" refers to "staggered" administration, such that the at least second therapeutic agent is administered to the subject at a different time from the administration of the protein construct of the invention. In general, the two agents are administered at a time interval effective to allow the two agents to exert their respective therapeutic effects, i.e., they are administered at a "biologically effective time interval". The at least second therapeutic agent may be administered to the subject at a biologically effective time before the administration of the protein of the invention or at a biologically effective time after the administration of the protein of the invention.

[0202] The in vivo methods and uses (e.g., therapeutic uses) described herein are generally carried out in mammals. Any mammal may be treated, for example, humans and farm animals, livestock, or laboratory animals. Specific examples include mice, rats, pigs, cats, dogs, sheep, rabbits, horses, cows, and monkeys (e.g., cynomolgus monkeys). However, the mammal is preferably a human. Thus, the term "patient" or "subject" as used herein includes such mammals. Thus, the subject or patient to be treated according to the present invention will preferably be a human.

[0203] Therapeutically effective amounts will be determined based on clinical evaluation and can be readily monitored.

[0204] The compositions and methods and uses of the invention may be used in conjunction with other therapeutic or diagnostic agents.

[0205] The present invention further comprises a kit, said kit comprising one or more protein constructs (or immunoconjugates) or compositions of the invention, or one or more nucleic acid molecules encoding the protein constructs of the invention, or one or more expression vectors, e.g. recombinant expression vectors, comprising the nucleic acid molecules of the invention, or one or more host cells comprising said expression vectors, e.g. recombinant expression vectors, or nucleic acid molecules of the invention. Preferably, said kit is for use in the methods and uses described herein, e.g. the therapeutic methods described herein, or for use in the in vitro assays or in vitro methods described herein. Preferably, said kit comprises instructions for use of the components of the kit. Preferably, said kit is for the treatment or prevention of a disease as described elsewhere herein, and optionally comprises instructions for use of the components of the kit for the treatment or prevention of such a disease.

[0206] The terms "a" and "an" are used throughout this application to mean "at least one," "at least a first," "one or more," or "a plurality" of the referenced component or step, unless an upper limit is specifically stated after the term.

[0207] Furthermore, when "comprise," "comprises," "has," or "having," or other equivalent terms are used herein, in some more specific embodiments these terms encompass "consists of" or "consists essentially of," or other equivalent terms. Methods comprising particular steps also include methods consisting of those steps, where appropriate.

[0208] The term "increase" or "enhance" or "improve" (or equivalent terms) as used herein includes any measurable increase or elevation when compared to a suitable control. Suitable controls would be easily identified by one skilled in the art, and suitable examples are described herein. The increase is considered significant, e.g., clinically or statistically significant, when compared to the level or value in a suitable control, and is preferably significant, e.g., clinically or statistically significant, at a probability value of less than or equal to 0.05.

[0209] The term "decrease" or "reduce" (or equivalent terms) as used herein includes a measurable decrease or decrease / reduction when compared to a suitable control. Suitable controls will be easily identified by those skilled in the art, and suitable examples are described herein. The decrease is significant, e.g., clinically or statistically significant, when compared to the level or value in a suitable control, and is preferably significant, e.g., clinically or statistically significant, with a probability value of less than or equal to 0.05.

[0210] Methods for determining the statistical significance of differences in the levels or values ​​of certain parameters are well known and documented in the art. For example, herein, a decrease or increase is generally considered to be statistically significant when a statistical comparison using a significance test such as Student's t-test, Mann-Whitney U-rank test, chi-square test or Fisher's exact test, one-way ANOVA test or two-way ANOVA test, as appropriate, shows a probability value of less than or equal to 0.05.

[0211] A partial list of the amino acid sequences disclosed herein and their sequence IDs (SEQ ID NOs): [Table 1-1] [Table 1-2] [Table 1-3] [Table 1-4]

[0212] Exemplary sequences of several IgA Fc-IgG Fc fusions for use in the extended constructs of the invention are shown below, all in the format IgG CH1, IgG hinge, IgA CH2, IgA CH3, IgG hinge, IgG CH1, IgG hinge, IgG CH2, IgG CH3. [Table 2] [Table 3] [Table 4] [Table 5] [Table 6] [Table 7] [Table 8] [Table 9] [Table 10] [Table 11] [Table 12]

[0213] Exemplary sequences of several IgA Fc-IgG Fc fusions for use in the extended constructs of the invention are shown below, all in the format IgA CH1, IgA hinge, IgA CH2, IgA CH3, IgG hinge, IgG CH2, IgG CH3. [Table 13] [Table 14]

[0214] The invention will now be further described in the following non-limiting Examples and with reference to the following figures. [Brief description of the drawings]

[0215] [Figure 1] Figure 1: Design of anti-Her2 antibody formats. Four antibody variants were produced based on IgG1 and IgA2. a) Tailless IgA2 has an Fc binding site for FcαRI. b) IgG1 has overlapping binding sites for FcγR and C1q in the lower hinge and CH2 domains, and a binding site for FcRn in the CH2-CH3 elbow region. The Fc engineered IgG1 variant contained a PGLALA substitution that abolished binding to low affinity FcγR and C1q, but not to FcRn. c) Tandem IgG1-IgA2 variant consists of full-length IgG1 fused to tailless, hinge-region-containing IgA2-Fc. d) Tandem IgA2-IgG1 variant consists of tailless, full-length IgA2 fused to hinge-region-containing IgG1-Fc. This figure was created with BioRender. [Diagram 2]Figure 2: pH-dependent human FcRn binding and extended plasma half-life. a) ELISA setup performed at pH 5.5 and pH 7.4, where titrated amounts of antibody variants were coated on wells, followed by addition of biotinylated human FcRn preincubated with ALP-conjugated streptavidin (STV). For clarity, only one tandem orientation is shown. This figure was generated with BioRender. b) ELISA results showing human FcRn binding to titrated amounts of antibody variants at pH 5.5 and pH 7.4. Shown as mean ± standard deviation of duplicates from a representative experiment. c) HERA results showing the amount of antibody variants detected in the medium by ELISA after a recycling step. Each antibody was added to HMEC-1 cells for 3 h, followed by extensive washing and incubation for another 3 h before medium was harvested. Shown as mean ± standard deviation of triplicates from a representative experiment. d) Plasma half-life study in human FcRn transgenic (Tg) mice, antibody formats were administered intravenously and blood samples were collected from day 1 to day 23. The figure was generated with BioRender. e) Plasma half-life (%) of antibody formats shown as mean ± SD (n=5). Two-tailed unpaired t-test, ns>0.05, ***=p<0.001. [Diagram 3] Figure 3: IgA2-IgG2 format and FcγRIIIb binding. a) IgG2 has distinct binding sites for FcγR, C1q, and FcRn. IgA2-IgG2 consists of full-length IgA2 without the tail fused to IgG2-Fc, including the hinge region. b) ELISA results performed at pH 7.4 showing binding of human FcγRIIIb to titrated amounts of antibody variants coated directly onto wells. Shown are the mean ± standard deviation of duplicates from a representative experiment. [Figure 4]Figure 4: Enhanced on-target C1q binding of IgA2-IgG1. a) ELISA setup used to measure C1q binding to Her2 capture antibody variants. Bound C1q was detected using rabbit anti-C1q antibody as primary antibody and anti-rabbit HRP-conjugated antibody as secondary antibody. b-d) ELISA results showing C1q binding to titrated amounts of anti-Her2 IgG-based antibody variants. Shown are duplicate mean ± standard deviation of a representative experiment. [Figure 5-1] Figure 5: Binding of Her2 antibody variants to FcγR and FcαRI. a) ELISA setup with recombinant Her2 coated onto wells followed by addition of titrating amounts of antibody formats. His6×-tagged human FcαRI or site-specific biotinylated human FcγR were added and detected with anti-His6×-ALP or ALP-conjugated streptavidin (STV), respectively. For clarity, only one tandem orientation is shown. This figure was created with BioRender. ELISA results for b) FcαRI, c) FcγRI, d) FcγRIIa-R131, e) FcγRIIa-H131, f) FcγRIIb, g) FcγRIIIa-F158, h) FcγRIIIa-V158, and i) FcγRIIIb at pH 7.4. Shown are the mean ± standard deviation of duplicates from a representative experiment. [Figure 5-2] Continued from Figure 5 [Figure 6-1] Figure 6: Fc receptor binding capacity of anti-Her2 IgG2-based antibody formats. ELISA results showing binding of a titration of the receptor to antibody variants: a) His6x-tagged human FcαRI, site-specifically biotinylated, b) FcγRI, c) FcγRIIa-R131, d) FcγRIIa-H131, e) FcγRIIb, f) FcγRIIIa-F158, g) FcγRIIIa-V158, and h) FcγRIIIb at pH 7.4. Shown are the mean ± standard deviation of duplicates from a representative experiment. [Figure 6-2] Figure 6 continued [Figure 7]Figure 7: Protein integrity and binding properties of UMAB2 IgG2-based antibody format. a) Non-reduced (NR) and reduced (R) SDS-PAGE analysis of affinity- and size-exclusion-purified antibodies. b-d) Antibody binding to anti-kappa, anti-Fc, and anti-IgA in ELISA. Shown are the mean ± standard deviation of duplicates from a representative experiment. [Figure 8] Figure 8: C1q binding and CDC activity in calcein AM-based assays. ELISA results showing C1q binding to titrated amounts of a) anti-CD20 and b) anti-Her2 antibody variants. c) Schematic of a CDC assay in which CD20-expressing target cells are loaded with calcein AM, then mixed with NHS and antibody variants, and cell lysis is measured by calcein release. For clarity, only one tandem orientation is shown. This figure was created with BioRender. d-e) Percent specific lysis of calcein-loaded CD20-expressing cell lines in the presence of antibody variants and NHS, using WSU-NHL and SU-DHL4 as target cells. Shown are the mean ± standard deviation of duplicates from a representative experiment. [Figure 9-1] Figure 9: Fc receptor binding capacity of anti-CD20 antibody formats. ELISA results showing binding of a titration of antibody variants of the receptor: a) His6x-tagged human FcαRI, site-specifically biotinylated, b) FcγRI, c) FcγRIIa-R131, d) FcγRIIa-H131, e) FcγRIIb, f) FcγRIIIa-F158, g) FcγRIIIa-V158, and h) FcγRIIIb at pH 7.4. Shown are the mean ± standard deviation of duplicates from a representative experiment. [Figure 9-2] Figure 9 continued [Figure 10-1]Figure 10: pH-dependent human FcRn binding of antibody variants. a-b) ELISA results showing human FcRn binding to titrating amounts of anti-Her2 antibody variants at pH 5.5 and pH 7.4. c-d) ELISA results showing human FcRn binding to titrating amounts of anti-CD20 antibody variants at pH 5.5 and pH 7.4. Shown are the mean ± standard deviation of duplicates from a representative experiment. [Figure 10-2] Continued from Figure 10 [Figure 11] Figure 11: CDC activity against Raji target cells in a calcein-AM based assay. a-b) Percentage of specific lysis of CD20-expressing Raji target cells in the presence of UMAB2 antibody variants and normal human serum (NHS). Shown are the mean ± standard deviation of duplicates from three independent experiments. [Figure 12] Figure 12: FcyRIIIa binding capacity of anti-Her2 antibody formats. ELISA results showing FcyRIIIa (V158 allotype) binding to Her2-specific IgA2-IgG1 and IgG1-IgA2 tandem formats. Shown are the mean ± standard deviation of duplicates from a representative experiment. [Figure 13] Figure 13: pH-dependent FcRn binding of anti-Her2 antibody variants. a-b) ELISA results showing human FcRn binding to titrating amounts of anti-Her2 antibody variants at pH 5.5 and pH 7.4. Shown are the mean ± standard deviation of duplicates from a representative experiment. [Figure 14] Figure 14: Plasma half-life of antibody formats in human FcRn-expressing mice in the presence of competition. Plasma half-life of antibody variants in human FcRn transgenic mice pretreated with 500mg / kg IVIg 2 days prior to administration of test antibody. Antibody variants were administered intravenously on day 0, followed by blood samples taken from days 1 to 23. Shown are the mean ± standard deviation (n=5). [Figure 15]Figure 15: Design of a novel IgG1 / IgA2 antibody format. Based on IgG1 and IgA2 specific for CD20 (UMAB2), two additional antibody variants were designed. The format comprises an IgG1 Fab2 fused to an IgA2 Fc (Cα2 and Cα3), followed by an IgG1 hinge, an IgG1 CH1, a third IgG1 hinge, and a C-terminal IgG1 Fc (CH2 and CH3). The C-terminal IgG1 Fc is either wild type (WT) or modified with amino acid substitutions by REW technology. [Figure 16] Figure 16: CDC activity of the novel IgG1 / IgA2 tandem format on Raji target cells in a calcein-AM based assay. Percent specific lysis of CD20 expressing Raji target cells in the presence of the novel UMAB2 IgG1 / IgA2 tandem format and NHS. Shown are the mean ± standard deviation of duplicates from a representative experiment. EXAMPLES

[0216] Example 1: Preparation of tandem IgA Fc-IgG Fc fusion protein Here, we report an antibody design panel that combines structural elements from IgA and IgG, specifically IgA Fc and IgG Fc, allowing efficient engagement of both FcγR and FcαRI. At the same time, we identified an IgA-IgG tandem format that is efficiently rescued from intracellular degradation by FcRn, leading to a long plasma half-life in human FcRn transgenic mice. This format combines full-length IgA2 with IgG1-Fc and was found to provide good induction of CDC-induced killing of CD20-expressing Raji cells and good pH-dependent binding to FcRn. In addition, when anti-CD20 IgA2-IgG2 was combined with an Fc engineering strategy for enhanced on-target hexamer formation, the tandem format produced a strong induction of CDC-induced killing of CD20-expressing cancer B cell lines.

[0217] material and method Antibody production and purification DNA segments encoding variable sequences from anti-HER2 trastuzumab or anti-CD20 UMAB2 (Meyer, S. et al. British journal of haematology 180, 808-820 (2018)) were subcloned in frame with IgA2-HC and kappa LC coding sequences into the previously described vectors pEE14.4-kappaLC, pEE14.4-IgA1, and pEE14.4-IgA2(m1). For the tandem variants, a cDNA encoding full-length IgG1-HC was subcloned into the hinge domain and the IgA2-C H 2-C H The cDNAs encoding the IgG1-Fc, IgG2-Fc, and IgG2-Fc-REW, including the corresponding hinge regions, were subcloned in frame with the DNA corresponding to the IgA2-HC for the IgA2-IgG1 / 2-Fc tandem variants. For the IgG1 Fab2-IgA2 Fc-H1-CH1-H1-IgG1 Fc extended tandem variants used in Example 2, the cDNAs encoding the entire constant region were synthesized and subcloned in frame with the trastuzumab or UMAB2 variable regions. The DNA segment encoding the tail of IgA2 was deleted. For the IgA2-IgG1 / 2-Fc tandem variants used in Example 2, the cDNAs encoding the entire constant region were synthesized and subcloned in frame with the trastuzumab or UMAB2 variable regions. Adherent HEK293E and Expi293 suspension cell lines were transiently co-transfected with the HC and LC encoding vectors, respectively, using Lipofectamine2000 or Expifectamine (ThermoFisher). The harvested supernatants were incubated with CaptureSelect IgG-C H1 was applied onto a pre-packed column (ThermoFisher) or onto a CaptureSelect IgA affinity matrix (ThermoFisher) packed in a column (Atoll) as described by the manufacturer. The eluted purified protein was buffer exchanged into phosphate buffered saline (PBS) (Merck), concentrated using an Amicon Ultra-15ml 100K spin column (Millipore), and then purified by size exclusion chromatography using a Superdex 200 increase 10 / 300GL column (GE Healthcare) coupled to an AEKTA Avant instrument (Cytiva Lifesciences). The eluted monomeric protein was concentrated using an Amicon Ultra-5ml 100K column (Millipore) and analyzed using SDS-PAGE (ThermoFisher).

[0218] ELISA ELISA was performed by coating EIA / RIA 96-well plates (Corning Costar) with antigen or antibody variants diluted in PBS (100 μl) and incubating overnight (ON) at 4°C. The next day, plates were blocked with 250 μl PBS (Merck) containing 4% skim milk (S) (ITW Reagent) for 1 h at room temperature (RT). Between all subsequent layers, plates were washed 4 times with 250 μl PBS containing 0.05% Tween 20 (T) (Merck). Unless otherwise stated, all samples were diluted in PBS / S / T in a total volume of 100 μl and incubated at room temperature for 1–2 h. To study antigen binding, recombinant human Her2 (1.0 μg / ml; Sino Biological) was coated onto wells and titrated amounts of antibody (7.0–0.0032 nM) were added the next day. ALP-conjugated anti-human kappa LC (Southern Biotech), anti-human IgG-Fc (Merck), or anti-IgA-Fc (Merck) antibodies were used as the second layer, where binding was visualized by adding ALP substrate (Merck) in diethanolamine buffer (pH 9.8). Absorbance was measured at 405 nm using a Sunrise spectrophotometer (TECAN). The same setup was used to measure binding of His6×-tagged human FcαRI (2 μg / ml; Sino Biological) and detected with ALP-conjugated His6×-tag antibody (Abcam) (1:5000). Additionally, binding to site-specific biotinylated FcγRI, FcγRIIa-H131, FcγRIIa-R131, FcγRIIb, FcγRIIIa-V156, FcγRIIIa-F156, and FcγRIIIb (0.25 μg / mL; Sino Biological) was performed by preincubating with ALP-conjugated streptavidin (Roche Diagnostic) (1:1 molar ratio) for 20 min at room temperature before addition to the plate. Bound proteins were detected as described above. Setup was also performed by directly coating the antibody variants (66.87 nM to 1.04 nM) followed by addition of biotinylated FcγRIIIb (0.5 μg / ml) preincubated with ALP-conjugated streptavidin (1:1 molar ratio). To study human FcRn binding, titrations of antibody formats (7.0 nM to 0.0032 nM) were coated onto wells. A 1:1 complex of site-specific biotinylated human soluble FcRn (0.25 μg / ml; Immunitrack) and ALP-conjugated streptavidin (Roche Diagnostics) was incubated for 20 min at room temperature before being added to the plate. Bound proteins were visualized as described above. The setup was performed using PBS / T and PBS / S / T at pH 5.5 or 7.4. To investigate human C1q binding, recombinant human Her2 (5.0 μg / ml or 10.0 μg / ml; Sino Biological) was coated and titrated antibody formats were added (42.0 nM-0.32 nM or 66.87 nM-1.04 nM). C1q binding was also performed by directly coating antibody variants (200.0 nM-3.12 nM). Human C1q (0.336 μg / mL or 0.5 μg / mL; Complement Technology) diluted in veronal buffer (Complement Technology) was added and incubated at 37°C for 30 min, followed by the addition of rabbit anti-C1q antibody (1:5000 or 1:10000; Dako). Finally, anti-rabbit-HRP (1:5000; GE Healthcare) was added to the plates and visualized by adding 3,3',5,5'-tetramethylbenzidine (TMB) substrate solution (Merck Millipore). The absorbance was measured at 620 nm or, after stopping the reaction by adding 50 μl of 1 M HCl, at 450 nm using a Sunrise spectrophotometer (TECAN).

[0219] HERA HERA was performed as previously described (Grevys et al., 2018, Nat. Commun.,9(1):621) with minor modifications. HMEC-1 cells stably expressing HA-hFcRn-EGFP (>100-fold) were cultured at 7.5 × 10 4The cells were seeded at 1 cell / well in 48-well plates (Corning Costar) and cultured for 1 day in growth medium (Gibco MCDB 131 medium (ThermoFisher), 10% heat-inactivated FCS (Merck), 2 mM L-glutamine (ThermoFisher), 1% Pencillin-Streptomycin (Merck), 10 ng / ml mouse epidermal growth factor (ThermoFisher), 1 μg / ml hydrocortisone (Merck)). Stable FcRn expression of the cells was ensured by adding 5 μg / ml blasticidin (ThermoFisher) and 100 μg / ml G418 (ThermoFisher). The cells were washed with Hank's Balanced Salt Solution (HBSS) (ThermoFisher) and starved for 1 hour before 800 nM of antibody format diluted in 125 μl HBSS (pH 7.4) was added to the cells and then incubated for 3 hours. The cells were then washed with ice-cold HBSS (pH 7.4) before growth medium without FCS but supplemented with MEM non-essential amino acids (ThermoFisher) was added to the cells and incubated for a further 3 hours. Media samples were then collected and the amount of antibody present was quantified by ELISA. This was done by coating with Her2 as described above followed by blocking (PBS / S) and washing (PBS / T) before adding the HERA sample. Binding of the antibody was detected using anti-Fc-ALP antibody (Merck) diluted in PBS / S / T (1:5000) and measured as described above.

[0220] CDC Calcein AM release CDC assay was performed. 1 × 10 7 WSU-NHL, SU-DHL4, or Raji cells were stained with 1 μl of calcein AM (Merck) for 30 min, washed with HBSS (Thermofisher), and resuspended in RPMI (Merck) at a cell density of 1 × 10 6 The cells were then resuspended at 5 × 10 4Cells were added to a V-shaped 96-well plate with titrations of anti-CD20 antibody (140 nM-0.729 nM) and NHS (Complement Technology) (final concentration 25%). RIPA buffer (Thermofisher) was used in place of antibody and NHS at a 1:1 dilution to measure maximum lysis of target cells. Basal release was measured in the absence of antibody and / or NHS. The plate was incubated at 37°C for 1 h and centrifuged at 2000 rpm for 10 min. Supernatants were transferred to a black clear-bottom optical 96-well Viewplate (Perkin Elmer) and fluorescence intensity was measured at 485 nm excitation / 510 nm emission using an Envision plate reader (Perkin Elmer) before calculating the percent lysis activity of each antibody format.

[0221] In vivo half-life studies Experiments were approved by The Jackson Laboratory's Animal Care and Use Committee and were performed at The Jackson Laboratory (JAX Services, Bar Harbor, ME) in accordance with the approved guidelines and regulations. Briefly, hemizygous Tg32 mice (B6.Cg-Fcgrt), which express human FcRn but not mouse FcRn, were cultured in vitro. tm1DcrPlasma half-life of antibody formats was determined using a Her2 specific antibody format (Her2+ / H2+Tg(FCGRT)32Dcr / DcrJ; The Jackson Laboratory). Where indicated, mice were pre-treated with 500 mg / kg intravenous immunoglobulin (IVIg) (Pirivigen; CSL Behring) 48 hours prior to administration of Her2 specific antibody formats. Each test article was studied in groups of 5 mice using only male mice (age: 8-9 weeks, body weight: 22-27 g). Mice were injected intravenously with equimolar amounts of test articles (IgA2 and IgG1 at 1.0 mg / kg, tandem variant at 1.3 mg / kg) and blood samples were collected from the retro-orbital sinus on days 1, 2, 3, 4, 5, 7, 10, 12, 16, 19, and 23 post-injection. Blood samples were processed immediately after collection, and plasma was isolated, diluted in 50% glycerol / PBS, and stored at -20°C. For ELISA quantification, plasma samples were diluted in PBS / S / T at dilutions of 1:50 to 1:400 (or 1:25, 1:50, 1:100, or 1:400). Half-life data are presented as percentages calculated based on the remaining protein at a given time point post-injection compared to day 1. Nonlinear regression analysis was performed using Prism 8 and half-life was calculated using the following formula:

number

[0222] statistical analysis Figure preparation and statistical analyses were performed using GraphPad Prism 8 for Mac (version 8.1.2; GraphPad Software Inc.), Microsoft Excel for Mac (version 16.35), and BioRender.com.

[0223] result Design and production of tandem antibodies To engineer the tandem antibody format, IgG1 and IgA2 heavy chain (HC)-encoding expression vectors with Her2 specificity from trastuzumab (Carter, P. et al. Proceedings of the National Academy of Sciences of the United States of America 89, 4285-4289 (1992)) were combined with sequences encoding IgA2-Fc (without tail) or IgG1-Fc, respectively, including the corresponding hinge sequence. In addition to wild-type IgG1, an engineered version (P329G, L234A and L235A; PGLALA) that abolished the ability to engage low affinity FcγR and C1q was also incorporated (Schlothauer, T. et al. Protein engineering, design & selection: PEDS 29, 457-466 (2016)). To produce the antibody, the constructed vector was combined with a vector encoding an anti-Her2 human kappa light chain (LC) (Meyer, S. et al. mAbs 8, 87-98 (2016) and Mester, S. et al. mAbs 13, 1893888 (2021)). Schematics of the antibody variants are shown in Figures 1a-1d. The purified proteins were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and the tandem variants migrated with a molecular weight approximately 50 kDa higher than that of the parental antibody. Under reducing conditions, the variants migrated in two distinct bands corresponding to their HC (50 kDa-75 kDa) and LC (25 kDa) bands, respectively (data not shown). The tandem variants were produced in amounts (1-2 g / L) similar to their parental counterparts and bound to recombinant human Her2 in enzyme-linked immunosorbent assays (ELISA) (data not shown).

[0224] The tandem design binds to FcRn and extends plasma half-life To compare the binding ability of the designed antibodies to human FcRn, ELISA was performed (Figure 2a).H 2-C H Since the tandem variants exhibit a strictly pH-dependent binding to the 3-elbow region at acidic pH and neither binding nor release at neutral pH (Fig. 1b), the assays were performed at pH 5.5 and 7.4, demonstrating that both tandem variants bind to the receptor with a pH dependence similar to that of IgG1, but not IgA2 (Fig. 2b). We next used adherent human microvascular endothelial cells (HMEC-1) stably overexpressing human FcRn to examine the ability of antibodies to rescue from intracellular degradation by FcRn using the human endothelial cell-based recycling assay (HERA) (Grevys, A. et al. Nature communications 9, 621 (2018)). Briefly, the antibodies were added to the cells, followed by an incubation step, and then the medium was removed and replaced. After further incubation, the medium was collected and the amount of antibody present was quantified as a measure of rescue from intracellular degradation. The HERA results showed that, as expected, IgG1 was rescued from intracellular degradation, but IgA2 was not. In addition, IgA2-IgG1 was rescued more efficiently than the reverse orientation tandem format of IgG1-IgA2, and even slightly more than IgG1 (Figure 2c). To address how this translates to plasma half-life in vivo, the antibody formats were administered intravenously to human FcRn transgenic mice, followed by blood samples taken over time and quantification by ELISA (Figure 2d). While IgA2 had a plasma half-life of only 0.9 days, IgA2-IgG1 showed the most dramatic increase in half-life (6.8 days), whereas IgG1-IgA2 had only a moderate half-life (2.7 days) compared to IgA2. Thus, a clear 2.5-fold difference was measured between the tandem variants, with IgA2-IgG1 showing the closest half-life to that of IgG1 (9.2 days) (Figure 2e).

[0225] Tandem variants exhibit Fc effector binding FcαRI is the C of IgA2 H 2-C H 3 elbow region (Herr, AB et al. Nature 423, 614-620 (2003)), whereas IgG1 binds to the C H The lower hinge and upper part of the 2 domains engage various FcγRs (Sondermann, P. et al. et al. Nature 406, 267-273 (2000) and Radaev, S et al. The Journal of biological chemistry 276, 16469-16477 (2001)) (Figure 1b). A set of ELISAs was performed to examine the Fc effector molecule binding properties of the tandem variants when captured by the cognate Her2 antigen (Figure 5a). The results revealed that FcαRI bound both tandem variants, albeit slightly weaker than IgA2, but not IgG1 (Figure 5b). When we investigated the binding to FcγR, we confirmed that all tandem variants showed similar binding reactions to IgG1, except that IgA2-IgG1 bound slightly stronger to FcγRIIb and FcγRIIIa-V, especially the decoy receptor FcγRIIIb (Figure 5c-Figure 5i). None of the various FcγRs, except for FcγRI, bound to the PGLALA-containing IgG1 variants, which is consistent with published data (Schlothauer, T. et al. Protein engineering, design & selection: PEDS 29, 457-466 (2016)). As expected, IgA2 did not bind to any of the FcγRs.

[0226] IgA2 in combination with IgG1- or IgG2-derived Fc exhibits Fc receptor binding Neutrophils express FcγRIIIb, which has been shown to act as a decoy receptor to interfere with IgG1-mediated ADCC (Brandsma, AM et al. Frontiers in immunology 10, 704 (2019), and Treffers, LW et al. Frontiers in Immunology 9, 3124 (2018)). However, the IgG2 subclass does not engage this particular receptor (Williams, TE et al. Biophys J 79, 1858-1866 (2000), and Bruhns, P. et al. Blood 113, 3716-3725 (2009)). Therefore, we redesigned the preferred tandem orientation, replacing IgG1-Fc with IgG2-Fc (Figure 3a). The generated anti-Her2 IgG2-based variants were successfully produced and bound recombinant Her2 (data not shown). With regard to Fc receptor binding, IgA2-IgG2 was again shown to have the ability to engage FcαRI, but again not to the same extent as IgA2. Meanwhile, both IgG2 and IgA2-IgG2 showed binding to FcγRIIa but not to FcγRIIIb (Figure 6). To confirm that the IgG2-based variants do not engage FcγRIIIb, ELISA wells were directly coated with a large amount of antibody and then the receptor was added. The results revealed a strict distinction: IgA2-IgG1 bound as well as IgG1, whereas IgG2, IgA2-IgG2, and IgA2 did not bind at all (Figure 3b).

[0227] IgA2-IgG1 shows enhanced C1q binding Complement factor C1q has a binding site that partially overlaps with FcγR, C HC1q binds IgG through the lower hinge and upper part of the C1q-2 domain (Figure 1b). However, the ability of C1q to initiate the complement cascade and induce CDC is dependent on IgG hexamer formation via Fc:Fc contacts upon binding to cell surface antigens (Diebolder, CA et al. Science (New York, NY) 343, 1260-1263 (2014)). To measure C1q binding of the designed antibody formats, they were captured on coated recombinant Her2 in an ELISA (Figure 4a). Remarkably, C1q bound more strongly to IgA2-IgG1 than to IgG1, and then to the IgG1-IgA2 tandem variant (Figures 4b-c). No or only weak C1q binding was measured for IgG1-PGLALA (Figures 4b-d), whereas no binding was observed for IgA2 (Figures 4b-c). Thus, fusing IgG1-Fc to IgA2 enhanced C1q binding, strongly supporting that this tandem design favors Fc:Fc interactions and hexamer formation upon binding to Her2. 69 binding was only weak (Fig. 4d), and the tandem IgA2-IgG2 variant did not engage C1q (Fig. 4d).

[0228] Fc engineering of IgA2-IgG2 enhances CDC activity To explore the effect of enhanced on-target C1q binding activity of the tandem format in a relevant system, we took advantage of a recently developed chimeric anti-CD20 specific antibody (UMAB2) (Evers, M. et al. mAbs 12, 1795505 (2020) and Meyer, S. et al. British journal of haematology 180, 808-820 (2018)) that can be used for CDC-induced killing of malignant CD20-expressing B cell lines. The variable sequences of this antibody were introduced into vectors encoding IgA2, IgG2, and the IgA2-IgG2 tandem variants. The rationale for this was to introduce three amino acid substitutions (Q311R, M428E, and N434W:REW) into the IgG2-based variants and evaluate their effect, despite the fact that IgG2 binds only weakly to C1q (Figure 4d). The anti-CD20 antibodies produced migrated at the expected molecular weight on SDS-PAGE and showed intact binding integrity by ELISA (Figure 7). In the absence of soluble recombinant CD20, antibody variants were directly coated and C1q binding was performed in ELISA, showing only weak binding to both IgG2 and IgA2-IgG2, but strongest binding to IgG1 (Figure 8a). This is consistent with the data for anti-Her2 variants when captured by antigen (Figure 4d) as well as the data for direct coating (Figure 8b). However, when the REW substitution was introduced, moderately enhanced C1q binding to IgG2 with both specificities was measured (Figures 8a-b). Similarly, IgA2-IgG2 showed weak binding, but stronger binding was measured when the REW substitution was introduced (Figures 8a-b). Nevertheless, the REW-containing IgA2-IgG2 tandem variants did not bind C1q as strongly as IgG1 (Figures 8a-b). Furthermore, we performed a calcein AM-based CDC assay, which relies on the labeling of target cells with the fluorescent dye calcein, released after antibody-mediated lysis, in the presence of NHS (Fig. 8c). Here, we used the CD20-high and CD20-low cell lines SU-DHL4 and WSU-NHL, respectively, and confirmed the strong CDC-inducing ability of IgA2-IgG2-REW, followed by IgA2-IgG2, at high CD20 expression (Fig. 8d-e). Although IgG2-REW showed a higher CDC induction than IgG2, the activity was lost at lower antibody concentrations, whereas the CDC activity of the tandem variants was maintained (Fig. 8d). At low CD20 expression, only the REW-containing variants induced CDC-mediated lysis (Fig. 8e). Taken together, our results demonstrate that REW substitutions enhance the CDC-inducing ability of IgG2, and this effect is even stronger when REW-containing IgG2-Fc is combined in tandem with full-length IgA2, which was observed in the presence of both high and low levels of CD20 expression. It was also confirmed that neither IgA2 nor IgA2-IgG2, even with REW substitutions, bound to the decoy receptor FcγRIIIb (Figure 9h), while retaining the ability to engage FcαRI (Figure 9a). There were no or only minor differences measured between wild-type and REW-containing variants, while the IgG2-based variants bound to FcγR more selectively than IgG1 (Figures 9a-h), consistent with the expected IgG2 binding properties (Williams, TE et al. Biophys J 79, 1858-1866 (2000), and Vidarsson, G. et al. Frontiers in Immunology 5 (2014)). Furthermore, REW-containing IgA2-IgG2 was shown to bind to human FcRn in a pH-dependent manner, with increased binding at acidic pH (Figure 10), strongly suggesting that the REW substitution leads to an extension of the half-life of the IgA2-IgG2 format.

[0229] Consideration As shown above, we have combined the structural features of IgA and IgG to generate an antibody format with multiple effector functions. We identified a format with a good ability to engage human FcRn. This was achieved with a tandem variant in which IgA2 was fused to IgG1-Fc, resulting in efficient rescue from degradation in HERA and a more than 7-fold longer half-life compared to IgA2 in human FcRn-expressing mice. This resulted in a 2.5-fold longer half-life than the reverse tandem orientation, IgG1-IgA2. Thus, we measured clear differences in the plasma half-life of the two tandem variants, indicating that the addition of IgG1-Fc to the C-terminal end of IgA2-HC is the most suitable orientation for efficient human FcRn-mediated rescue from degradation both in vitro and in vivo. Furthermore, we investigated the ability of this format to engage C1q and induce CDC-mediated cell killing. In particular, we found that the IgA2-IgG1 orientation advantageously improved C1q binding compared to IgG1 when captured on Her2 in an ELISA. We also investigated the IgA2-IgG2 format in combination with the REW substitution. Advantageously, the REW substitution was found to improve binding to C1q in ELISA when directly coated (anti-Her2 and anti-CD20) for both IgG2 and IgA2-IgG2. Similar to the effect observed in the C1q binding assay, IgG2-REW acquired the ability to mediate CDC, which was most pronounced at high target concentrations. REW-containing IgA2-IgG2 performed extremely well in cellular CDC assays when targeting CD20-expressing cancer B cells. This suggests that IgA2-IgG2-REW has a high ability to form Fc:Fc hexameric structures upon binding to CD20 on target cells, allowing for a much more efficient induction of the complement cascade than IgG2-REW. This result also strongly suggests that REW substitution leads to an extension of the half-life of the IgAFc-IgGFc format. Consistent with this, REW-containing IgA2-IgG2 bound to human FcRn in a pH-dependent manner, with increased binding at acidic pH. Thus, REW substitution was shown to further enhance the properties of the construct, such as C1q binding and CDC, as well as FcRn binding.

[0230] Further results Strong CDC activity due to IgG1-Fc at the C-terminal end To investigate the CDC activity of the tandem orientation, we performed a calcein AM release CDC assay (Figure 8c), which relies on labeling of target cells with the fluorescent dye calcein AM, which is released after antibody-mediated lysis in the presence of NHS. Here, we used Raji cells, which express low levels of CD20 relative to the levels of complement inhibitory receptors and are therefore considered "hard-to-kill". These results (Figure 11a) confirmed the potent CDC-inducing ability of our IgA2-IgG1 construct and were consistent with efficient C1q binding similar to that of IgG1-WT in ELISA, whereas the reversed tandem orientation IgG1-IgA2 format showed poor binding, similar to IgG1-PGLALA, a mutant that abolished C1q binding.

[0231] Fc engineering enhances CDC activity We showed that CDC activity could be further enhanced by introducing a REW mutation into the tandem format (Fig. 11b). Specifically, the IgA2-IgG1-REW orientation induced a higher level of CDC compared to the reverse tandem orientation. By introducing REW into the IgA2-IgG1 construct, we achieved a somewhat higher level of CDC activity than that of IgG1-WT. Taken together, the IgA2-IgG1 orientation demonstrated superiority in engaging C1q and mediating CDC upon binding to target antigen compared to the IgG1-IgA2 orientation. In other experiments, we also demonstrated that neither tandem format elicited complement activation in human serum in the absence of target antigen binding (data not shown), which is important from a safety perspective if the constructs are to be used therapeutically.

[0232] Strong binding of IgA2-IgG1 to FcyRIIIa NK cells express FcyRIIIa and are believed to be the major cell population mediating ADCC by cross-linking of IgG antibodies. Therefore, we investigated the binding of FcyRIIIa in tandem orientation when captured by the cognate Her2 antigen. The results (see Figure 12) showed that while both orientations / formats bound to the receptor, the IgA2-IgG1 orientation showed stronger binding compared to the reverse tandem orientation. Taken together, this indicates that the IgA2-IgG1 orientation may have improved binding ability to FcyRIIIa on NK cells compared with the opposite orientation, and thus may mediate ADCC efficiently.

[0233] The tandem format binds to FcRn ELISA was performed to measure the binding ability of the tandem format to human FcRn (Figure 2a). FcRn is the C domain of IgG1. H 2-C HSince the 3-elbow region exhibits strictly pH-dependent binding at acidic pH and neither binding nor release at neutral pH (Figure 1b), the assay was performed at pH 5.5 and pH 7.4. The results (Figure 13) demonstrated that both tandem variants exhibit pH-dependent binding to the receptor, but the IgA2-IgG1 orientation exhibited stronger binding than the reverse orientation and also stronger binding than IgG1 WT. When REW was introduced into these formats, enhanced binding was measured at pH 5.5, but binding remained low at pH 7.4.

[0234] Tandem format extends half-life in competitive models To measure the in vivo plasma half-life of the antibody formats in a physiologically relevant context, the antibodies were administered intravenously to human FcRn transgenic mice preloaded with IVIg, and blood samples were subsequently collected over time and quantified by ELISA (Figure 2d). In addition to the reduced ability of mouse IgG to bind human FcRn in the transgenic mice, which have low levels of endogenous mouse IgG as a result of their pathogen-free housing environment, we injected the mice with human IgG to mimic a physiologically relevant situation in which high levels of IgG compete for FcRn binding and recycling. The plasma half-life of IgA2 was only 0.8 days (Figure 14), whereas an increased half-life was observed for IgA2-IgG1 (3.1 days). However, only a slightly longer half-life was measured for IgG1 with the same specificity (4.4 days). When the REW mutation was introduced into the tandem format (IgA2-IgG1), a half-life similar to that of IgG1 was measured (4.5 days). Taken together, the IgA2-IgG1 tandem format, in contrast to IgA2 alone, is processed and rescued by FcRn in the presence of high levels of human IgG, which competes for FcRn engagement. The IgA2-IgG1 variant has a reduced half-life compared to the parental IgG1, but this is compensated for by the introduction of the REW.

[0235] Example 2: Design of an alternative (extended) IgA Fc-IgG Fc tandem format Here we report an alternative tandem design that retains structural elements from IgA Fc and IgG Fc, specifically the combination of IgA2 Fc and IgG1 Fc, but allows efficient engagement to both FcγR and FcαRI, similar to Example 1. This format combines IgA Fc (here IgA2 Fc) with IgG Fc (here IgG1 Fc), further separated by the inclusion of an additional IgG1-CH1 domain (i.e., an extended format). Specifically, the IgG1 Fab and hinge regions are fused to the IgA2 Fc, followed by an IgG1 hinge, an IgG1 CH1 domain, a third IgG1 hinge, and a full-length IgG1 Fc (such a construct is designated IgG1 Fab2-IgA2 Fc-H1-CH1-H1-IgG1 Fc, see FIG. 15). This design was further combined with the REW Fc engineering strategy, where the REW substitutions were introduced into the C-terminal IgG1 Fc domain. Anti-CD20 and anti-HER2 specificity is provided by an antigen-binding fragment in the form of an IgG1 Fab2 fragment.

[0236] material and method Antibody production and purification are described in Example 1. CDC assays were performed as described in Example 1.

[0237] result Anti-CD20 IgG1 Fab2-IgA2 Fc-H1-CH1-H1-IgG1 Fc (WT or REW) was examined for its ability to mediate CDC against "hard to kill" target cells in the form of Raji cells expressing low levels of CD20, and the new tandem format resulted in stronger killing activity than that of the IgA2-IgG1 format as well as IgG1 (Figure 16). Introduction of the REW substitution into the C-terminal IgG1 Fc domain further enhanced CDC activity. These new extended CH-1 containing formats were also shown to bind to Fcγ receptors when tested on FcγRIIIa-V, similar to the IgA2-IgG1 containing tandem constructs of the invention described in Example 1 and IgG1-WT (data not shown).

Claims

1. A protein construct comprising a human IgA Fc region and a human IgG Fc region, wherein the C-terminus of the human IgA Fc region is linked to the N-terminus of the human IgG Fc region.

2. The protein construct of claim 1 , wherein the IgA Fc region is an IgA1 Fc region or an IgA2 Fc region, preferably an IgA2 Fc region.

3. The protein construct of claim 1 , wherein the IgG Fc region is an IgG1 Fc region or an IgG2 Fc region.

4. The protein construct of claim 1 , wherein the construct is capable of binding to an Fc receptor or capable of Fc effector function.

5. 5. The protein construct of claim 4, wherein the construct is capable of binding to FcαR, capable of binding to one or more FcγRs, capable of binding to FcRn, capable of binding to C1q, capable of inducing CDC, capable of inducing ADCC, and / or capable of inducing ADCP.

6. The protein construct of claim 1 , wherein the C-terminus of the human IgA Fc region is linked to the N-terminus of the human IgG Fc region by a linker and / or by a CH1 domain.

7. The protein construct of claim 6 , wherein the linker is a peptide linker, preferably a peptide linker comprising an antibody hinge region.

8. The protein construct of claim 1 , wherein the IgG Fc region or the IgA Fc region comprises a mutation or modification that enhances effector function or increases plasma half-life.

9. 9. The protein construct of claim 8, wherein the mutation or modification increases binding to an Fc receptor, preferably increases binding to FcRn or increases C1q binding.

10. 9. The protein construct of claim 8, wherein the Fc region is a modified IgG Fc region comprising: (i) an arginine (R) residue or a similar residue, such as a lysine (K) residue, at position 311 (or a position corresponding thereto); (ii) a glutamic acid (E) residue or a similar residue, such as an aspartic acid (D) residue, at position 428 (or a position corresponding thereto); and (iii) at position 434 (or a position corresponding thereto), a tryptophan (W) residue or a similar residue such as a tyrosine (Y) or phenylalanine (F) residue;

11. 2. The protein construct of claim 1, further comprising a targeting domain, preferably a receptor domain or receptor ligand, or an antigen-binding domain.

12. The protein construct of claim 11 , wherein the antigen-binding domain is an antibody or an antigen-binding fragment thereof.

13. The protein construct of claim 12, wherein the antibody is an IgA antibody or an antigen-binding fragment thereof, or an IgG antibody or an antigen-binding fragment thereof.

14. 12. The protein construct of claim 11, wherein the targeting domain binds to a target molecule on a cancer cell or an infectious pathogen.

15. 15. The protein construct of claim 14, wherein the cancer cells are derived from a solid tumor or a blood cancer, or the infectious pathogen is a bacterium.

16. One or more nucleic acid molecules comprising a nucleotide sequence encoding the protein construct of claim 1; or one or more expression vectors comprising such nucleic acid molecules; or One or more host cells comprising the expression vector, the nucleic acid molecule, or the protein construct of claim 1.

17. 16. A method for producing a protein construct according to any one of claims 1 to 15, comprising the steps of:

17. A method comprising the steps of: (i) culturing a host cell comprising one or more expression vectors or one or more nucleic acid sequences according to claim 16 under conditions suitable for expression of the encoded protein construct, and optionally (ii) isolating or obtaining the expressed protein construct from said host cell or from the growth medium / supernatant.

18. 18. The method of claim 17, further comprising purifying the protein product and / or formulating the protein product into a composition comprising at least one additional component.

19. 17. A composition, preferably a pharmaceutically acceptable composition, comprising a protein construct according to any one of claims 1 to 15, or one or more nucleic acid molecules or expression vectors according to claim 16.

20. 16. A protein construct according to any one of claims 1 to 15 for use in therapy.

21. 21. A protein construct for use according to claim 20, comprising: the protein construct comprises a targeting domain, preferably a receptor domain or receptor ligand, or an antigen-binding domain, which binds to a target or antigen; the protein construct is for use in the treatment or prevention of a disease characterized by expression of the target or antigen; Preferably, said disease is cancer or an infectious disease caused by a pathogen expressing said target or said antigen.

22. Use of a protein construct according to any one of claims 1 to 15, in the manufacture of a medicament for use in the treatment or prevention of a disease characterized by expression of said target or antigen, wherein the protein construct comprises a targeting domain, preferably a receptor domain or receptor ligand, or an antigen-binding domain, which binds to said target or antigen, Preferably, the disease is cancer or an infectious disease caused by a pathogen expressing said target or said antigen.

Images