Antibody construct
Multivalent antibody constructs with high-fidelity pairing and single-step purification address stability and efficiency issues in multispecific antibody production, enabling effective high-throughput screening and therapeutic applications.
Patent Information
- Application Number
- JP2022162426
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2017-09-07
- Filing Date
- 2022-10-07
- Publication Date
- 2024-11-27
- Estimated Expiration
- 2037-10-18
AI Technical Summary
Current methods for producing multispecific antibodies face challenges such as poor stability, inefficient expression, and complex purification processes, limiting their application in high-throughput screening and therapeutic settings.
Development of multivalent antibody constructs with high-fidelity pairing of cognate polypeptide chains, allowing for easy expression using conventional systems and single-step purification with CH1 affinity resins, enhancing stability and suitability for high-throughput screening.
The constructs achieve high-fidelity assembly, efficient expression, and purification, making them suitable for high-throughput screening and effective therapeutic agents with long-term stability.
Smart Images

Figure 0007591015000020 
Figure 0007591015000021 
Figure 0007591015000022
Abstract
Description
[Technical field]
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims the benefit under 35 U.S.C. 119(e) of co-pending U.S. Provisional Patent Application No. 62 / 555,498, filed September 7, 2017, U.S. Provisional Patent Application No. 62 / 549,894, filed August 24, 2017, and U.S. Provisional Patent Application No. 62 / 410,054, filed October 19, 2016, the disclosures of which are incorporated herein by reference in their entireties.
[0002] 2. Sequence Listing This application contains a Sequence Listing, which was submitted via EFS-Web and is incorporated herein by reference in its entirety. The ASCII copy was created on X / X / 2017, is named XXXXXUS_sequencelisting.txt, and is X,XXX,XXX bytes in size. [Background technology]
[0003] 3.Background Antibodies are invaluable tools in the medical field. The importance of monoclonal antibodies, particularly including their role in scientific research and medical diagnostics, has been widely recognized for decades. However, the full potential of antibodies, particularly their successful use as therapeutic agents, has only recently been demonstrated, as shown by the success of the therapies adalimumab (Humira), rituximab (Rituxan), infliximab (Remicade), bevacizumab (Avastin), trastuzumab (Herceptin), pembrolizumab (Keytruda), and ipilimumab (Yervoy). Following these clinical successes, interest in antibody therapeutics will likely only increase. Thus, there is a need in the art for efficient generation and production of antibodies, both in research drug development and downstream in clinical settings.
[0004] In the field of antibody therapy, an area of active research is the design and use of multispecific antibodies, i.e., a single antibody engineered to recognize multiple targets.These antibodies offer the promise of greater therapeutic control.For example, there is a need to improve target specificity to reduce the off-target effects associated with many antibody therapies, especially antibody-based immunotherapy.In addition, multispecific antibodies provide new therapeutic strategies, such as synergistic targeting of multiple cell receptors, especially in the context of immunotherapy.
[0005] Despite the promise of multispecific antibodies, their production and use have suffered from numerous constraints that limit their practical application. In general, all multispecific antibody platforms must solve the problem of ensuring high-fidelity pairing between pairs of cognate heavy and light chains. However, many problems exist across the various platforms. For example, antibody chain engineering can lead to poor stability of the assembled antibody, poor expression and folding of the antibody chains, and / or the generation of immunogenic peptides. Other approaches have issues with impractical manufacturing processes, such as complex in vitro assembly reactions or purification methods. In addition, some platforms have issues with the inability to simply and efficiently plug in different antibody binding domains. These various issues associated with multispecific antibody production limit the application of many platforms, especially their use in high-throughput screening, which is necessary for many therapeutic drug pipelines.
[0006] Thus, there is a need for antibody platforms that allow high-level expression and efficient purification, and in particular, for multispecific antibody platforms that improve the ability to produce multispecific antibodies that have direct application in both research and therapeutic settings. Summary of the Invention [Means for solving the problem]
[0007] 4. Overview The inventors have designed a variety of novel multivalent antibody constructs. The configuration of these multivalent binding molecules drives high-fidelity pairing of cognate polypeptide chains that together form antigen-binding sites in monospecific, bispecific, trispecific, and tetraspecific constructs. The binding molecules can be easily expressed using conventional antibody expression systems, including in vitro cell-free translation systems and mammalian transient transfection systems, and purified in a single step using CH1 affinity resin. The high-fidelity assembly, high levels of in vitro expression, and single-step purification of expression products make these constructs well suited for high-throughput screening of libraries of variable regions. These constructs also exhibit long-term stability, making them well suited as multispecific therapeutic agents.
[0008] Thus, in a first aspect, there is provided a binding molecule comprising a first and a second polypeptide chain, (a) the first polypeptide chain comprises Domain A, Domain B, Domain D, and Domain E, where the domains are arranged from N-terminus to C-terminus in an ABDE orientation, and Domain A has a VL amino acid sequence, Domain B has a CH3 amino acid sequence, Domain D has a CH2 amino acid sequence, and Domain E has a constant region domain amino acid sequence; (b) the second polypeptide chain comprises Domain F and Domain G, where the domains are arranged from N-terminus to C-terminus in an FG orientation, and Domain F has a VH amino acid sequence, and Domain G has a CH3 amino acid sequence; and (c) the first and second polypeptides associate via interactions between the A and F domains and between the B and G domains to form the binding molecule.
[0009] In certain embodiments, the binding molecule further comprises a third and fourth polypeptide chain, (a) the third polypeptide chain comprises Domain H, Domain I, Domain J and Domain K, where the domains are arranged N-terminally to C-terminally in an HIJK orientation, Domain H has a variable region domain amino acid sequence, Domain I has a constant region domain amino acid sequence, Domain J has a CH2 amino acid sequence, and Domain K has a constant region domain amino acid sequence; (b) the fourth polypeptide chain comprises Domain L and Domain M, where the domains are arranged N-terminally to C-terminally in an LM orientation, Domain L has a variable region domain amino acid sequence, and Domain M has a constant region domain amino acid sequence; (c) the third and fourth polypeptides associate through interactions between the H and L domains and between the I and M domains; (d) the first and third polypeptides associate through interactions between the D and J domains and between the E and K domains to form the binding molecule.
[0010] In certain embodiments, the amino acid sequences of the B and G domains are identical, where the sequence is an endogenous CH3 sequence.
[0011] In certain embodiments, the amino acid sequences of the B and G domains are different and each separately comprises an orthogonal modification in the endogenous CH3 sequence, where the B domain interacts with the G domain and neither the B domain nor the G domain significantly interacts with the CH3 domain lacking the orthogonal modification. In certain embodiments, the orthogonal modification comprises a mutation that creates an engineered disulfide bridge between domains B and G. In certain embodiments, the mutation that creates an engineered disulfide bridge is a S354C mutation in one of the B and G domains and a 349C mutation in the other domain. In certain embodiments, the orthogonal modification comprises a knob-in-hole mutation. In certain embodiments, the knob-in-hole mutation is a T366W mutation in one of the B and G domains and a T366S, L368A and Y407V mutation in the other domain. In certain embodiments, the orthogonal modification comprises a charge pair mutation. In certain embodiments, the charge pair mutations are a T366K mutation in one of the B and G domains and a L351D mutation in the other domain.
[0012] In certain embodiments, domain E has a CH3 amino acid sequence. In certain embodiments, the amino acid sequences of the E and K domains are identical, where the sequence is an endogenous CH3 sequence.
[0013] In certain embodiments, the amino acid sequences of the E and K domains are different. In certain embodiments, the different sequences each separately comprise an orthogonal modification to the endogenous CH3 sequence, where the E domain interacts with the K domain, and neither the E domain nor the K domain significantly interacts with the CH3 domain lacking the orthogonal modification. In certain embodiments, the orthogonal modification comprises a mutation that creates an engineered disulfide bridge between domains E and K. In certain embodiments, the mutation that creates an engineered disulfide bridge is a S354C mutation in one of the E and K domains, and a 349C mutation in the other domain. In certain embodiments, the orthogonal modification in the E and K domains comprises a knob-in-hole mutation. In certain embodiments, the knob-in-hole mutation is a T366W mutation in one of the E and K domains, and a T366S, L368A, and Y407V mutation in the other domain. In certain embodiments, the orthogonal modification comprises a charge pair mutation. In certain embodiments, the charge pair mutation is a T366K mutation in either the E or K domain and a corresponding L351D mutation in the other domain.
[0014] In certain embodiments, the amino acid sequences of the E domain and the K domain are the endogenous sequences of two different antibody domains that are selected to have a specific interaction that promotes the specific association between the first polypeptide and the third polypeptide.In certain embodiments, the two different amino acid sequences are a CH1 sequence and a CL sequence.In certain embodiments, the domain I has a CL sequence and the domain M has a CH1 sequence.
[0015] In certain embodiments, domain H has a VL sequence and domain L has a VH sequence.
[0016] In certain embodiments, domain H has a VL amino acid sequence; domain I has a CL amino acid sequence; domain K has a CH3 amino acid sequence; domain L has a VH amino acid sequence; and domain M has a CH1 amino acid sequence.
[0017] In certain embodiments, the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, and the interaction between the H and L domains forms a second antigen-binding site specific for a second antigen.
[0018] In certain embodiments, the binding molecule further comprises a fifth polypeptide chain, wherein (a) the first polypeptide chain further comprises domain N and domain O, wherein the domains are arranged from N-terminus to C-terminus in a NOABDE orientation, domain N has a VL amino acid sequence, and domain O has a CH3 amino acid sequence; (b) the binding molecule further comprises a fifth polypeptide chain, wherein the fifth polypeptide chain comprises domain P and domain Q, wherein the domains are arranged from N-terminus to C-terminus in a PQ orientation, domain P has a VH amino acid sequence, and domain Q has a CH3 amino acid sequence; (c) the first and fifth polypeptides associate via interactions between the N and P domains and between the O and Q domains to form the binding molecule.
[0019] In certain embodiments, (a) the amino acid sequences of Domain N and Domain A are identical, the amino acid sequence of Domain H is different from the amino acid sequences of Domain N and A, the amino acid sequences of Domain O and Domain B are identical, the amino acid sequence of Domain I is different from the amino acid sequences of Domain O and B, the amino acid sequences of Domain P and Domain F are identical, the amino acid sequence of Domain L is different from the amino acid sequences of Domain P and F, the amino acid sequences of Domain Q and Domain G are identical, and the amino acid sequence of Domain M is different from the amino acid sequences of Domain Q and G; (b) the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, the interaction between the H and L domains forms a second antigen-binding site specific for a second antigen, and Domain N and Domain P form a third antigen-binding site specific for the first antigen.
[0020] In certain embodiments, (a) the amino acid sequences of Domain N, Domain A, and Domain H are different, the amino acid sequences of Domain O, Domain B, and Domain I are different, the amino acid sequences of Domain P, Domain F, and Domain L are different, and the amino acid sequences of Domain Q, Domain G, and Domain M are different; (b) the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, the interaction between the H and L domains forms a second antigen-binding site specific for a second antigen, and Domain N and Domain P form a third antigen-binding site specific for a third antigen.
[0021] In certain embodiments, the binding molecule further comprises a sixth polypeptide chain, wherein (a) the third polypeptide chain further comprises a domain R and a domain S, wherein the domains are arranged from N-terminus to C-terminus in an orientation of RSHIJK, where domain R has a VL amino acid sequence, and domain S has a constant domain amino acid sequence; (b) the binding molecule further comprises a sixth polypeptide chain, wherein the sixth polypeptide chain comprises a domain T and a domain U, wherein the domains are arranged from N-terminus to C-terminus in an orientation of TU, where domain T has a VH amino acid sequence, and domain U has a constant domain amino acid sequence; (c) the third and sixth polypeptides associate via interactions between the R and T domains and between the S and U domains to form the binding molecule.
[0022] In certain embodiments, (a) the amino acid sequences of Domain R and Domain A are identical, the amino acid sequence of Domain H is different from the amino acid sequences of Domain R and A, the amino acid sequences of Domain S and Domain B are identical, the amino acid sequence of Domain I is different from the amino acid sequences of Domain S and B, the amino acid sequences of Domain T and Domain F are identical, the amino acid sequence of Domain L is different from the amino acid sequences of Domain T and F, the amino acid sequences of Domain U and Domain G are identical, and the amino acid sequence of Domain M is different from the amino acid sequences of Domain U and Domain G; (b) the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, the interaction between the H and L domains forms a second antigen-binding site specific for a second antigen, and Domain R and Domain T form a third antigen-binding site specific for the first antigen.
[0023] In certain embodiments, (a) the amino acid sequences of Domain R and Domain H are identical, the amino acid sequence of Domain A differs from the amino acid sequences of Domain R and H, the amino acid sequences of Domain S and Domain I are identical, the amino acid sequence of Domain B differs from the amino acid sequences of Domain S and I, the amino acid sequences of Domain T and Domain L are identical, the amino acid sequence of Domain F differs from the amino acid sequences of Domain T and L, the amino acid sequences of Domain U and Domain M are identical, and the amino acid sequence of Domain G differs from the amino acid sequences of Domain U and M; (b) the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, the interaction between the H and L domains forms a second antigen-binding site specific for a second antigen, and Domain R and Domain T form a third antigen-binding site specific for the second antigen.
[0024] In certain embodiments, (a) the amino acid sequences of Domain R, Domain A, and Domain H are different, the amino acid sequences of Domain S, Domain B, and Domain I are different, the amino acid sequences of Domain T, Domain F, and Domain L are different, and the amino acid sequences of Domain U, Domain G, and Domain M are different; (b) the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, the interaction between the H and L domains forms a second antigen-binding site specific for a second antigen, and Domain R and Domain T form a third antigen-binding site specific for a third antigen.
[0025] In certain embodiments, the binding molecule further comprises a fifth and sixth polypeptide chain, (a) the first polypeptide chain further comprises domain N and domain O, where the domains are arranged from N-terminus to C-terminus in a NOABDE orientation; (b) the third polypeptide chain further comprises domain R and domain S, where the domains are arranged from N-terminus to C-terminus in a RSHIJK orientation; (c) the binding molecule further comprises a fifth and sixth polypeptide chain, where the fifth polypeptide chain comprises domain P and domain Q, where the domains are arranged from N-terminus to C-terminus in a PQ orientation, and the sixth polypeptide chain comprises domain T and domain U, where the domains are arranged from N-terminus to C-terminus in a TU orientation; (d) the first and fifth polypeptides associate through interactions between the N and P domains and between the O and Q domains, and the third and sixth polypeptides associate through interactions between the R and T domains and between the S and U domains to form the binding molecule.
[0026] In certain embodiments, (a) the amino acid sequences of Domain N and Domain A are identical, the amino acid sequences of Domain H and Domain R are identical, the amino acid sequences of Domain O and Domain B are identical, the amino acid sequences of Domain I and Domain S are identical, the amino acid sequences of Domain P and Domain F are identical, the amino acid sequences of Domain L and Domain T are identical, the amino acid sequences of Domain Q and Domain G are identical, and the amino acid sequences of Domain M and Domain U are identical; (b) the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, Domain N and Domain P forms a second antigen-binding site specific for the first antigen, the interaction between the H and L domains forms a third antigen-binding site specific for a second antigen, and the interaction between the R and T domains forms a fourth antigen-binding site specific for the second antigen.
[0027] In a particular embodiment, the binding molecule of claim 36, wherein: (a) the amino acid sequences of Domain H and Domain A are identical, the amino acid sequences of Domain N and Domain R are identical, the amino acid sequences of Domain I and Domain B are identical, the amino acid sequences of Domain O and Domain S are identical, the amino acid sequences of Domain L and Domain F are identical, the amino acid sequences of Domain P and Domain T are identical, the amino acid sequences of Domain M and Domain G are identical, and the amino acid sequences of Domain Q and Domain U are identical; (b) the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, Domain N and Domain P forms a second antigen-binding site specific for a second antigen, the interaction between the H and L domains forms a third antigen-binding site specific for the first antigen, and the interaction between the R and T domains forms a fourth antigen-binding site specific for the second antigen.
[0028] In certain embodiments, the sequence forming the junction between the A and B domains is IKRTPREP or IKRTVREP, In certain embodiments, the sequence forming the junction between the F and G domains is SSASPREP.
[0029] In certain embodiments, at least one CH3 amino acid sequence has a C-terminal tripeptide insertion linking the CH3 amino acid sequence and the hinge amino acid sequence, wherein the tripeptide insertion is selected from the group consisting of PGK, KSC and GEC.
[0030] In certain embodiments, the sequence is a human sequence. In certain embodiments, at least one CH3 amino acid sequence is an IgG sequence. In certain embodiments, the IgG sequence is an IgG1 sequence. In certain embodiments, at least one CH3 amino acid sequence has one or more isoallotypic mutations. In certain embodiments, the isoallotypic mutations are D356E and L358M. In certain embodiments, the CL amino acid sequence is C カッパ It is an array.
[0031] Also provided herein is a binding molecule comprising first, second, third and fourth polypeptide chains, wherein (a) the first polypeptide chain comprises Domain A, Domain B, Domain D and Domain E, wherein the domains are arranged from N-terminus to C-terminus in an ABDE orientation, and Domain A has a VL amino acid sequence, Domain B has a CH3 amino acid sequence, Domain D has a CH2 amino acid sequence and Domain E has a constant region domain amino acid sequence; (b) the second polypeptide chain comprises Domain F and Domain G, wherein the domains are arranged from N-terminus to C-terminus in an FG orientation, and Domain F has a VH amino acid sequence and Domain G has a CH3 amino acid sequence; (c) the third polypeptide chain comprises Domain H, Domain I, Domain J and Domain K, wherein the domains are arranged from N-terminus to C-terminus in an HIJK orientation, and Domain H has a variable region domain amino acid sequence and Domain (d) a fourth polypeptide chain comprises domain L and domain M, where the domains are arranged N-terminus to C-terminus in an LM orientation, domain L having a variable region domain amino acid sequence and domain M having a constant region domain amino acid sequence; (e) the first and second polypeptides associate through an interaction between the A domain and the F domain and an interaction between the B domain and the G domain; (f) the third and fourth polypeptides associate through an interaction between the H domain and the L domain and an interaction between the I domain and the M domain; and (g) the first and third polypeptides associate through an interaction between the D domain and the J domain and an interaction between the E domain and the K domain to form the binding molecule.
[0032] In certain embodiments, domain E has a CH3 amino acid sequence; domain H has a VL amino acid sequence; domain I has a CL amino acid sequence, domain K has a CH3 amino acid sequence; domain L has a VH amino acid sequence; and domain M has a CH1 amino acid sequence.
[0033] Also disclosed herein are pharmaceutical compositions comprising any of the binding molecules disclosed or described herein and a pharma- ceutically acceptable carrier.
[0034] Also disclosed herein are methods of treatment comprising administering to a subject in need of treatment a pharmaceutical composition comprising any of the binding molecules disclosed or described herein and any of the pharma- ceutically acceptable carriers disclosed or described herein. 5. Brief description of the drawings [Brief description of the drawings]
[0035] [Figure 1] Figure 1 shows the alignment of the CH3-CH3 IgG1 dimer pair paired with CH1-CL. The quaternary structures are aligned with an RMSD of approximately 1.6 Å2.
[0036] [Diagram 2] FIG. 2 shows a schematic diagram of the construction of the various binding molecules (also referred to as antibody constructs) described herein, along with their respective naming conventions.
[0037] [Diagram 3] FIG. 3 shows a higher resolution schematic of the polypeptide chains and their domains, along with their respective nomenclature conventions, for the bivalent 1×1 antibody constructs described herein.
[0038] [Figure 4] FIG. 4 shows the configuration of an exemplary bivalent monospecific construct.
[0039] [Diagram 5]FIG. 5 shows data from a biolayer interferometry (BLI) experiment described in Example 1, in which bivalent monospecific binding molecules with the configuration shown in FIG. 4 [polypeptide 1: VL-CH3 (knob)-CH2-CH3 / polypeptide 2: VH-CH3 (hole)] were assayed. The antigen binding site was specific for TNFα. BLI responses from binding molecule immobilization and TNFα binding to the immobilized construct indicate robust and specific bivalent binding to the antigen. The data are consistent with the molecules having a high percentage of intended pairing with polypeptide 1 and polypeptide 2.
[0040] [Figure 6] FIG. 6 shows the characteristics of an exemplary bivalent 1×1 bispecific binding molecule, “BC1.”
[0041] [Figure 7] Figure 7A shows a size-exclusion chromatography (SEC) analysis of "BC1" demonstrating that a single-step CH1 affinity purification step (CaptureSelect™ CH1 affinity resin) yields a single monodisperse peak of >98% non-aggregated bivalent protein via gel filtration. Figure 7B shows comparative literature data from SEC analysis of CrossMab bivalent antibody constructs [data from Schaefer et al., (Proc Natl Acad Sci USA., July 5, 2011; 108(27):11187-92)].
[0042] [Figure 8] Figure 8A is a cation exchange chromatographic elution profile of "BC1" after one-step purification using CaptureSelect™ CH1 affinity resin, showing a single tight peak. Figure 8B is a cation exchange chromatographic elution profile of "BC1" after purification using standard Protein A purification.
[0043] [Figure 9]FIG. 9 shows a non-reducing SDS-PAGE gel of "BC1" at various stages of purification.
[0044] [Figure 10] Figures 10A and 10B show SDS-PAGE gels of "BC1" after single-step CH1-affinity purification under both non-reducing and reducing conditions (Figure 10A) compared with SDS-PAGE gels of CrossMab bispecific antibodies under non-reducing and reducing conditions published in the literature (Figure 10B).
[0045] [Figure 11] Figures 11A and 11B show mass spectrometry analysis of "BC1" demonstrating two distinct heavy chains (Figure 11A), and two distinct light chains (Figure 11B) under reducing conditions.
[0046] [Figure 12] FIG. 12 shows mass spectrometry analysis of "BC1" purified under non-reducing conditions, confirming the absence of imperfect pairing after purification.
[0047] [Figure 13] FIG. 13 shows accelerated stability study data demonstrating the stability of "BC1" over 8 weeks at 40° C. compared to two IgG control antibodies.
[0048] [Figure 14] FIG. 14 shows the characteristics of an exemplary bivalent 1×1 bispecific binding molecule “BC6,” which is further described in Example 3.
[0049] [Figure 15] Figure 15A shows a size-exclusion chromatography (SEC) analysis of "BC6" after one-step purification using CaptureSelect™ CH1 affinity resin, demonstrating that the single-step CH1 affinity purification results in a single monodisperse peak and the absence of non-covalent aggregates. Figure 15B shows an SDS-PAGE gel of "BC6" under non-reducing conditions.
[0050] [Figure 16] FIG. 16 shows the characteristics of an exemplary bivalent, bispecific binding molecule “BC28,” which is further described in Example 4.
[0051] [Figure 17] FIG. 17 shows SDS-PAGE analysis under non-reducing conditions after single-step CH1 affinity purification of "BC28", "BC29", "BC30", "BC31" and "BC32".
[0052] [Figure 18] FIG. 18 shows the SEC analysis of "BC28" and "BC30," respectively, after one-step purification using CaptureSelect™ CH1 affinity resin.
[0053] [Figure 19] FIG. 19 shows the characteristics of an exemplary bivalent, bispecific binding molecule “BC44,” which is further described in Example 5.
[0054] [Figure 20] Figures 20A and 20B show size exclusion chromatography data for two bivalent binding molecules, "BC15" and "BC16," respectively, under accelerated stability testing conditions. "BC15" and "BC16" have different variable region-CH3 junctions.
[0055] [Figure 21] FIG. 21 shows a schematic diagram of the polypeptide chains and their domains, along with their respective nomenclature, for the trivalent 2×1 antibody constructs described herein.
[0056] [Figure 22] FIG. 22 shows the characteristics of an exemplary trivalent 2×1 bispecific binding molecule “BC1-2×1,” which is further described in Example 7.
[0057] [Figure 23]FIG. 23 shows a non-reducing SDS-PAGE of "BC1" and "BC1-2x1" proteins expressed using the ThermoFisher Expi293 transient transfection system at various stages of purification.
[0058] [Figure 24] FIG. 24 compares the avidity of the bivalent 1×1 construct "BC1" with the avidity of the trivalent 2×1 construct "BC1-2×1" using Octet (Pall ForteBio) biolayer interference analysis.
[0059] [Diagram 25] FIG. 25 shows the salient features of the trivalent 2×1 construct “TB111.”
[0060] [Figure 26] FIG. 26 shows a schematic diagram of the polypeptide chains and their domains, along with their respective nomenclature, for the trivalent 1×2 antibody constructs described herein.
[0061] [Figure 27] FIG. 27 shows characteristics of an exemplary trivalent 1×2 construct “CTLA4-4×Nivo×CTLA4-4,” which is further described in Example 10.
[0062] [Figure 28] FIG. 28 is an SDS-PAGE gel in which the lanes showing the trivalent 1×2 construct “CTLA4-4×Nivo×CTLA4-4” construct under non-reducing (“−DTT”) and reducing (“+DTT”) conditions are boxed.
[0063] [Figure 29] Figure 29 shows a comparison of antigen binding between two antibodies: the bivalent 1x1 construct "CTLA4-4xOX40-8" and the trivalent 1x2 construct "CTLA4-4xNivoxCTLA4-4." "CTLA4-4xOX40-8" binds monovalently to CTLA4, while "CTLA4-4xNivoxCTLA4-4" binds bivalently to CTLA4.
[0064] [Diagram 30] FIG. 30 shows characteristics of an exemplary trivalent 1×2 trispecific construct “BC28-1×1×1a,” which is further described in Example 11.
[0065] [Diagram 31] FIG. 31 shows size exclusion chromatography of "BC28-1x1x1a" after transient expression and single-step CH1 affinity resin purification showing a single clearly defined peak.
[0066] [Diagram 32] FIG. 32 shows the results of SDS-PAGE under non-reducing and reducing conditions of bivalent and trivalent constructs after transient expression and one-step purification using CaptureSelect™ CH1 affinity resin, respectively, as further described in Example 12.
[0067] [Diagram 33] Figures 33A-33C show Octet binding analysis to three antigens: PD1, antigen "A" and CTLA4. As further described in Example 13, Figure 33A shows the binding of "BC1" to PD1 and antigen "A"; Figure 33B shows the binding of the bivalent bispecific construct "CTLA4-4xOX40-8" to CTLA4, antigen "A" and PD1; Figure 33C shows the binding of the trivalent trispecific "BC28-1x1x1a" to PD1, antigen "A" and CTLA4.
[0068] [Diagram 34] FIG. 34 shows a schematic diagram of the polypeptide chains and their domains, along with their respective nomenclature, for certain tetravalent 2×2 constructs described herein.
[0069] [Diagram 35] FIG. 35 illustrates certain salient features of an exemplary tetravalent 2×2 construct, “BC22-2×2,” which is further described in Example 14.
[0070] [Diagram 36] FIG. 36 is a non-reducing SDS-PAGE gel comparing the 2×2 tetravalent "BC22-2×2" construct with the 1×2 trivalent "BC12-1×2" and 2×1 trivalent "BC21-2×1" constructs at various stages of purification.
[0071] [Figure 37] FIG. 37 shows the configuration of an exemplary tetravalent 2×2 construct.
[0072] [Figure 38] FIG. 38 shows a schematic diagram of the polypeptide chains and their domains, along with their respective nomenclature, for certain tetravalent constructs described herein.
[0073] [Figure 39] FIG. 39 shows an exemplary configuration of a bispecific tetravalent construct.
[0074] [Diagram 40] FIG. 40 shows an exemplary configuration for a trispecific tetravalent construct utilizing a common light chain strategy.
[0075] [Diagram 41] Figure 41 shows bispecific antigen engagement by the tetravalent construct outlined in Figure 39, demonstrating that this construct was capable of co-engagement. Biolayer Interference (BLI) responses from B-Body immobilization and TNFα binding to the immobilized construct are consistent with the molecule having a high percentage of intended chain pairing.
[0076] [Diagram 42] Figure 42 shows flow cytometry analysis of binding of B-Body to cell surface antigens. Shaded signals indicate cells without antigen; dotted signals indicate cells transiently transfected with surface antigen.
[0077] [Diagram 43] FIG. 43 shows an exemplary configuration of a trivalent construct.
[0078] [Diagram 44] FIG. 44 shows an exemplary configuration of a trivalent construct.
[0079] [Diagram 45] FIG. 45 shows SDS-PAGE results under non-reducing and reducing conditions of bivalent and trivalent constructs after transient expression and one-step purification using CaptureSelect™ CH1 affinity resin, respectively, as further described in Example 17.
[0080] [Diagram 46] FIG. 46 shows the differences in thermal transitions of "BC24jv," "BC26jv," and "BC28jv," measured to assess the pairing stability of the junction variants. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0081] The drawings depict various embodiments of the present invention for purposes of illustration only. Those skilled in the art will readily recognize from the following description that alternative embodiments of the structures and methods illustrated herein may be utilized without departing from the principles of the present invention as described herein.
[0082] 6. Detailed Description 6.1.Definition Unless otherwise defined, all technical and scientific terms used herein have the meaning commonly understood by one of ordinary skill in the art to which this invention belongs. As used herein, the following terms have the meanings given below.
[0083] "Antigen-binding site" means the region of a binding molecule that specifically recognizes or binds to a given antigen or epitope.
[0084] "B-Body" means any of the binding molecule constructs described herein.
[0085] As used herein, the term "treat" or "treatment" refers to both therapeutic treatment and prophylactic or preventative measures, where the objective is to prevent or slow down (alleviate) undesirable physiological changes or disorders, such as the progression of multiple sclerosis, arthritis, or cancer. Beneficial or desired clinical results include, but are not limited to, alleviation of symptoms, whether detectable or undetectable, reduction in the extent of disease, stable (i.e., not worsening) state of disease, delay or slowing of disease progression, amelioration or alleviation of disease state, and remission (partial or complete remission). "Treatment" can also mean prolonging survival compared to expected survival in the absence of treatment. Subjects in need of treatment include those already suffering from a condition or disorder, as well as those prone to having a condition or disorder, or those in whom a condition or disorder is to be prevented.
[0086] "Subject," "individual," "animal," "patient," or "mammal" means any subject for which diagnosis, prevention, or treatment is desired, particularly a mammalian subject. Mammalian subjects include humans, domestic animals, farm animals, as well as zoo, sport, or pet animals, such as dogs, cats, guinea pigs, rabbits, rats, mice, horses, cattle, dairy cows, and the like.
[0087] The term "sufficient amount" means an amount sufficient to produce a desired effect, for example an amount sufficient to modulate protein aggregation in a cell.
[0088] A "therapeutically effective amount" is an amount effective for ameliorating symptoms of a disease. Just as prevention can be considered treatment, a therapeutically effective amount can be a "prophylactically effective amount."
[0089] 6.2. Other Rules of Interpretation Unless otherwise specified, all references to sequences herein are to amino acid sequences.
[0090] Unless otherwise specified, antibody constant region residue numbering is according to the Eu index as set forth in www.imgt.org / IMGTScientificChart / Numbering / Hu_IGHGnber.html#refs (accessed August 22, 2017) and Edelman et al., Proc. Natl. Acad. USA, 63:78-85 (1969), which are incorporated herein by reference in their entirety, and identifies residues according to their position in the endogenous constant region sequence, regardless of the physical location of the residue within the chain of the binding molecules described herein. "Endogenous sequence" or "native sequence" refers to any sequence, including both nucleic acid and amino acid sequences, that is derived from an organism, tissue, or cell and has not been artificially modified or mutated.
[0091] In this disclosure, the words "comprises," "comprising," "containing," "having," "includes," "including," and linguistic variations thereof have the meaning ascribed to them under U.S. patent law and permit the presence of additional components other than those expressly recited.
[0092] Ranges provided herein are understood to be shorthand for all values within the range, including the recited endpoints. For example, a range of 1 to 50 is understood to include any number, combination of numbers, or subrange from the group consisting of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, and 50.
[0093] As used herein, unless otherwise stated or clear from context, the term "or" is understood to be inclusive. Unless otherwise stated or clear from context, the terms "a," "an," and "the" are understood to be singular or plural.
[0094] Unless otherwise specified or clear from the context, the term "about" as used herein is understood to be within the range normally accepted in the art, for example, within 2 standard deviations of the mean. About can be understood to be within 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1%, 0.5%, 0.1%, 0.05% or 0.01% of the indicated value. Unless otherwise clear from the context, all numerical values provided herein are modified by the term about.
[0095] 6.3. Binding molecules In a first aspect, a binding molecule is provided.
[0096] With reference to FIG. 3 , in a first set of embodiments, the binding molecule comprises first and second polypeptide chains, (a) the first polypeptide chain comprises Domain A, Domain B, Domain D, and Domain E, where the domains are arranged, from N-terminus to C-terminus, in an ABDE orientation, and Domain A has a VL amino acid sequence, Domain B has a CH3 amino acid sequence, Domain D has a CH2 amino acid sequence, and Domain E has a constant region domain amino acid sequence; (b) the second polypeptide chain comprises Domain F and Domain G, where the domains are arranged, from N-terminus to C-terminus, in an FG orientation, and Domain F has a VH amino acid sequence and Domain G has a CH3 amino acid sequence; (c) the first and second polypeptides associate via interactions between the A and F domains and between the B and G domains to form the binding molecule.
[0097] Also referring to FIG. 3 , in a second set of embodiments, the binding molecule further comprises a third and a fourth polypeptide chain, (a) the third polypeptide chain comprises Domain H, Domain I, Domain J and Domain K, where the domains are arranged N-terminally to C-terminally in an HIJK orientation, and Domain H has a variable region domain amino acid sequence, Domain I has a constant region domain amino acid sequence, Domain J has a CH2 amino acid sequence, and Domain K has a constant region domain amino acid sequence; (b) the fourth polypeptide chain comprises Domain L and Domain M, where the domains are arranged N-terminally to C-terminally in an LM orientation, and Domain L has a variable region domain amino acid sequence and Domain M has a constant region domain amino acid sequence; (c) the third and fourth polypeptides associate via interactions between the H and L domains and between the I and M domains; and (d) the first and third polypeptides associate via interactions between the D and J domains and between the E and K domains to form the binding molecule.
[0098] In various embodiments, the first and third polypeptide chains are identical to each other in sequence, and the second and fourth polypeptide chains are identical to each other in sequence, In these embodiments, association of the first and third polypeptide chains via an interaction between Domain E and Domain K (see Section 6.3.15 below) forms a bivalent monospecific antibody construct, such as the construct exemplified in Example 1 below.
[0099] In other embodiments, the first and third polypeptide chains are non-identical to each other in sequence, and the second and fourth polypeptide chains are non-identical to each other in sequence. In these embodiments, association of the first and third polypeptide chains via an interaction between Domain E and Domain K (see Section 6.3.15 below) can form a bivalent, bispecific antibody construct.
[0100] Domain A (VL) In the binding molecules described herein, domain A has a VL amino acid sequence. The VL amino acid sequence useful in the binding molecules described herein is an antibody light chain variable domain sequence. In typical sequences in both natural antibodies and antibody constructs described herein, a specific VL amino acid sequence associates with a specific VH amino acid sequence to form an antigen-binding site. In various embodiments, the VL amino acid sequence is a mammalian sequence, including a human sequence, a synthetic sequence, or a combination of a human sequence, a non-human mammalian sequence, a mammalian sequence, and / or a synthetic sequence, as described in more detail in Sections 6.3.1.1 and 6.3.1.2 below.
[0101] In various embodiments, the VL amino acid sequence is a mutant sequence of a naturally occurring sequence. In certain embodiments, the VL amino acid sequence is a lambda (λ) light chain variable domain sequence. In certain embodiments, the VL amino acid sequence is a kappa (κ) light chain variable domain sequence. In a preferred embodiment, the VL amino acid sequence is a kappa (κ) light chain variable domain sequence.
[0102] In the binding molecules described herein, the C-terminus of Domain A is linked to the N-terminus of Domain B. In certain embodiments, Domain A has a VL amino acid sequence that is mutated at its C-terminus at the junction between Domain A and Domain B, as described in further detail in Section 6.3.19.1 and Example 6 below.
[0103] Complementarity-Determining Regions The VL amino acid sequence contains highly variable sequences called "complementarity determining regions" (CDRs), typically three CDRs (CDR1, CD2, and CDR3). In various embodiments, the CDRs are mammalian sequences, including but not limited to mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences. In preferred embodiments, the CDRs are human sequences. In various embodiments, the CDRs are naturally occurring sequences. In various embodiments, the CDRs are naturally occurring sequences that have been mutated to alter the binding affinity of the antigen binding site for a particular antigen or epitope. In certain embodiments, the naturally occurring CDRs have been mutated in vivo in a host through affinity maturation and somatic hypermutation. In certain embodiments, the CDRs have been mutated in vitro through methods including but not limited to PCR mutagenesis and chemical mutagenesis. In various embodiments, the CDRs are synthetic sequences, including but not limited to CDRs obtained from random sequence CDR libraries and rationally designed CDR libraries.
[0104] 6.3.1.2. Framework Regions and CDR Grafting The VL amino acid sequence comprises "framework region" (FR) sequences. FRs are generally conserved sequence regions (see Section 6.3.1.1.) that act as scaffolds for the interspersed CDRs, typically in the FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4 configuration (N-terminus to C-terminus). In various embodiments, the FRs are mammalian sequences, including but not limited to mouse, rat, hamster, rabbit, camel, donkey, goat, and human sequences. In preferred embodiments, the FRs are human sequences. In various embodiments, the FRs are naturally occurring sequences. In various embodiments, the FRs are synthetic sequences, including but not limited to rationally designed sequences.
[0105] In various embodiments, the FR and CDR are both derived from the same naturally occurring variable domain sequence. In various embodiments, the FR and CDR are derived from different variable domain sequences, where the CDRs are grafted onto the FR scaffold, and the CDRs provide specificity for a particular antigen. In certain embodiments, the grafted CDRs are all derived from the same naturally occurring variable domain sequence. In certain embodiments, the grafted CDRs are derived from different variable domain sequences. In certain embodiments, the grafted CDRs are synthetic sequences, including but not limited to CDRs obtained from random sequence CDR libraries and rationally designed CDR libraries. In certain embodiments, the grafted CDRs and FRs are derived from the same species. In certain embodiments, the grafted CDRs and FRs are derived from different species. In preferred grafted CDR embodiments, the antibody is "humanized", where the grafted CDRs are non-human mammalian sequences, including but not limited to mouse, rat, hamster, rabbit, camel, donkey and goat sequences, and the FRs are human sequences. Humanized antibodies are discussed in more detail in U.S. Patent No. 6,407,213, which is incorporated by reference in its entirety for all that it teaches. In various embodiments, portions or specific sequences of FRs from one species are used to replace portions or specific sequences of FRs from another species.
[0106] Domain B (CH3) In the binding molecule, domain B is a CH3 amino acid sequence, which is the sequence of the C-terminal domain of an antibody heavy chain, as described herein.
[0107] In various embodiments, the CH3 sequence is a mammalian sequence, including but not limited to mouse, rat, hamster, rabbit, camel, donkey, goat and human sequence.In a preferred embodiment, the CH3 sequence is a human sequence.In a particular embodiment, the CH3 sequence is from IgA1, IgA2, IgD, IgE, IgM, IgG1, IgG2, IgG3, IgG4 isotype, or is a CH4 sequence from IgE or IgM isotype.In a preferred embodiment, the CH3 sequence is from IgG1 isotype.
[0108] In certain embodiments, the CH3 sequence is an endogenous sequence. In certain embodiments, the CH3 sequence is UniProt Accession No. P01857 amino acids 224-330. In various embodiments, the CH3 sequence is a segment of an endogenous CH3 sequence. In certain embodiments, the CH3 sequence has an endogenous CH3 sequence that lacks N-terminal amino acids G224 and Q225. In certain embodiments, the CH3 sequence has an endogenous CH3 sequence that lacks C-terminal amino acids P328, G329 and K330. In certain embodiments, the CH3 sequence has an endogenous CH3 sequence that lacks N-terminal amino acids G224 and Q225 and C-terminal amino acids P328, G329 and K330. In preferred embodiments, the binding molecule has multiple domains with CH3 sequences, where the CH3 sequence can refer to both full-length endogenous CH3 sequences and CH3 sequences that lack N-terminal amino acids, C-terminal amino acids, or both.
[0109] In certain embodiments, the CH3 sequence is an endogenous sequence having one or more mutations. In certain embodiments, the mutations are one or more orthogonal mutations introduced into the endogenous CH3 sequence to guide the specific pairing of a particular CH3 sequence, as described in more detail below in Sections 6.3.14.1-6.3.14.3.
[0110] In certain embodiments, the CH3 sequence is engineered to reduce the immunogenicity of the antibody by replacing certain amino acids of one allotype with amino acids of another allotype, referred to herein as isoallotypic mutations, as described in more detail in Stickler et al. (Genes Immun., April 2011; vol. 12(3):213-221), which is incorporated herein by reference for all its teachings. In certain embodiments, certain amino acids of the G1m1 allotype are replaced. In a preferred embodiment, the isoallotypic mutations D356E and L358M are made in the CH3 sequence.
[0111] In a preferred embodiment, domain B has a human IgG1 CH3 amino acid sequence with the following mutational changes: P343V; Y349C; and a tripeptide insert 445P, 446G, 447K. In another preferred embodiment, domain B has a human IgG1 CH3 sequence with the following mutational changes: T366K; and a tripeptide insert 445K, 446S, 447C. In yet another preferred embodiment, domain B has a human IgG1 CH3 sequence with the following mutational changes: Y349C and a tripeptide insert 445P, 446G, 447K.
[0112] In certain embodiments, domain B has a human IgG1 CH3 sequence with, differently, a 447C mutation incorporated into the endogenous CH3 sequence.
[0113] In the binding molecules described herein, the N-terminus of Domain B is linked to the C-terminus of Domain A. In certain embodiments, Domain B has a CH3 amino acid sequence that is mutated at its N-terminus at the junction between Domain A and Domain B, as described in further detail in Section 6.3.19.1 and Example 6 below.
[0114] In the binding molecule, the C-terminus of Domain B is linked to the N-terminus of Domain D. In certain embodiments, Domain B has a CH3 amino acid sequence extended C-terminally to the junction between Domain B and Domain D, as described in further detail in Section 6.3.19.3 below.
[0115] Domain D (CH2) In the binding molecules described herein, domain D has a CH2 amino acid sequence. The CH2 amino acid sequence is the CH2 amino acid sequence of the third domain of a natural antibody heavy chain, as described herein, from N-terminus to C-terminus. In various embodiments, the CH2 sequence is a mammalian sequence, including but not limited to mouse, rat, hamster, rabbit, camel, donkey, goat and human sequence. In a preferred embodiment, the CH2 sequence is a human sequence. In certain embodiments, the CH2 sequence is derived from an IgA1, IgA2, IgD, IgE, IgG1, IgG2, IgG3, IgG4 or IgM isotype. In a preferred embodiment, the CH2 sequence is derived from an IgG1 isotype.
[0116] In certain embodiments, the CH2 sequence is an endogenous sequence. In certain embodiments, the sequence is amino acids 111-223 of UniProt Accession No. P01857. In preferred embodiments, the CH2 sequence has an N-terminal hinge region peptide that connects the N-terminal variable domain-constant domain segment to the CH2 domain, as discussed in more detail below in Section 6.3.19.3.
[0117] In the binding molecule, the N-terminus of Domain D is linked to the C-terminus of Domain B. In certain embodiments, Domain B has a CH3 amino acid sequence that is extended at the C-terminus at the junction between Domain D and Domain B, as described in further detail below in Section 6.3.19.3.
[0118] In the binding molecule, the C-terminus of domain D is linked to the N-terminus of domain E. In certain embodiments, domain D is linked to the N-terminus of domain E, which has a CH1 amino acid sequence or a CL amino acid sequence, as described in more detail in Section 6.3.19.5, below.
[0119] 6.3.4. Domain E (Constant Region) In the binding molecule, domain E has a constant region domain amino acid sequence. As described herein, a constant region domain amino acid sequence is the sequence of an antibody constant region domain.
[0120] In various embodiments, the constant region sequence is a mammalian sequence, including but not limited to mouse, rat, hamster, rabbit, camel, donkey, goat and human sequences. In preferred embodiments, the constant region sequence is a human sequence. In certain embodiments, the constant region sequence is from an antibody light chain. In certain embodiments, the constant region sequence is from a lambda or kappa light chain. In certain embodiments, the constant region sequence is from an antibody heavy chain. In certain embodiments, the constant region sequence is an antibody heavy chain sequence that is an IgA1, IgA2, IgD, IgE, IgG1, IgG2, IgG3, IgG4 or IgM isotype. In preferred embodiments, the constant region sequence is from an IgG1 isotype. In certain embodiments, the constant region sequence is a CH3 sequence. CH3 sequences are described in more detail in section 6.3.2 above.
[0121] In certain embodiments, the constant region sequence is mutated to include one or more orthogonal mutations. In preferred embodiments, domain E has a constant region sequence that is a CH3 sequence that includes either a knob-hole (synonymously "knobs-in-holes", "KIH") orthogonal mutation, as described in more detail below in Section 6.3.14.2, as well as an S354C or Y349C mutation that forms an engineered disulfide bridge with the CH3 domain containing the orthogonal mutation, as described in more detail below in Section 6.3.14.1. In some preferred embodiments, the knob-hole orthogonal mutation is a T366W mutation.
[0122] In certain embodiments, the constant region domain sequence is a CH1 sequence. CH1 sequences are described in more detail in Section 6.3.4.1 below. In certain embodiments, the N-terminus of the CH1 domain is linked to the C-terminus of the CH2 domain, as described in more detail in Section 6.3.19.5 below.
[0123] In certain embodiments, the constant region sequence is a CL sequence. CL sequences are described in more detail below in section 6.3.4.2. In certain embodiments, the N-terminus of the CL domain is linked to the C-terminus of the CH2 domain, as described in more detail below in 6.3.19.5.
[0124] CH1 domain The CH1 amino acid sequence, as described herein, is the sequence of the second domain of the antibody heavy chain, referring to the N-terminus to the C-terminus. In certain embodiments, the CH1 sequence is an endogenous sequence. In various embodiments, the CH1 sequence is a mammalian sequence, including but not limited to mouse, rat, hamster, rabbit, camel, donkey, goat and human sequence. In a preferred embodiment, the CH1 sequence is a human sequence. In certain embodiments, the CH1 sequence is derived from an IgA1, IgA2, IgD, IgE, IgG1, IgG2, IgG3, IgG4 or IgM isotype. In a preferred embodiment, the CH1 sequence is derived from an IgG1 isotype. In a preferred embodiment, the CH1 sequence is UniProt Accession No. P01857 amino acids 1-98.
[0125] 6.3.4.2.CL Domain The CL amino acid sequence useful in the binding molecule described herein is an antibody light chain constant domain sequence.In certain embodiments, the CL sequence is an endogenous sequence.In various embodiments, the CL sequence is a mammalian sequence, including but not limited to mouse, rat, hamster, rabbit, camel, donkey, goat and human sequence.In a preferred embodiment, the CL sequence is a human sequence.
[0126] In certain embodiments, the CL amino acid sequence is a lambda (λ) light chain constant domain sequence. In certain embodiments, the CL amino acid sequence is a human lambda light chain constant domain sequence. In a preferred embodiment, the lambda (λ) light chain sequence is UniProt Accession No. P0CG04.
[0127] In certain embodiments, the CL amino acid sequence is a kappa (κ) light chain constant domain sequence. In a preferred embodiment, the CL amino acid sequence is a human kappa (κ) light chain constant domain sequence. In a preferred embodiment, the kappa light chain sequence is UniProt Accession No. P01834.
[0128] Domain F (VH) In the binding molecules, domain F has a VH amino acid sequence. The VH amino acid sequence in the binding molecules described herein is an antibody heavy chain variable domain sequence. In a typical antibody configuration, both in natural and in the binding molecules described herein, a particular VH amino acid sequence associates with a particular VL amino acid sequence to form an antigen-binding site. In various embodiments, the VH amino acid sequence is a mammalian sequence, including a human sequence, a synthetic sequence, or a combination of non-human mammalian sequences, mammalian sequences and / or synthetic sequences, as described in more detail in Sections 6.3.1.1 and 6.3.1.2 above. In various embodiments, the VH amino acid sequence is a mutant sequence of a naturally occurring sequence.
[0129] Domain G (CH3) In the binding molecule, domain G has a CH3 amino acid sequence, which is described in further detail in Section 6.3.2 above.
[0130] In certain preferred embodiments, domain G has a human IgG1 CH3 sequence with the following mutational changes: S354C; and a tripeptide insert 445P, 446G, 447K. In some preferred embodiments, domain G has a human IgG1 CH3 sequence with the following mutational changes: S354C; and a tripeptide insert 445P, 446G, 447K. In some preferred embodiments, domain G has a human IgG1 CH3 sequence with the following mutational changes: L351D; and a tripeptide insert 445G, 446E, 447C.
[0131] 6.3.7. Domain H (variable region) In the binding molecule, domain H has a variable region domain amino acid sequence. The variable region domain amino acid sequence is an antibody variable region domain amino acid sequence, including VL and VH antibody domain sequences, as described herein. VL and VH sequences are described in more detail above in sections 6.3.1 and 6.3.5, respectively. In a preferred embodiment, domain H has a VL antibody domain sequence.
[0132] 6.3.8. Domain I (Constant Region) In the binding molecules described herein, domain I has a constant region domain amino acid sequence. The constant region domain amino acid sequence is described in more detail in section 6.3.4 above. In a preferred embodiment, domain I has a constant region sequence that is CL from a kappa light chain, as discussed in more detail in section 6.3.4.2.
[0133] Domain J (CH2) In the binding molecule, Domain J has a CH2 amino acid sequence, which is described in more detail in Section 6.3.3, supra. In a preferred embodiment, the CH2 amino acid sequence has an N-terminal hinge region linking Domain J to Domain I, as described in more detail in Section 6.3.19.4, below.
[0134] In the binding molecule, the C-terminus of domain J is linked to the N-terminus of domain K. In certain embodiments, domain J is linked to the N-terminus of domain K having a CH1 amino acid sequence or a CL amino acid sequence, as described in further detail in Section 6.3.19.5, below.
[0135] 6.3.10. Domain K (Constant Region) In the binding molecule, domain K has a constant region domain amino acid sequence. The constant region domain amino acid sequence is described in more detail in section 6.3.4 above. In a preferred embodiment, domain K has a constant region sequence that is a CH3 sequence that contains either a S354C or Y349C mutation that forms an engineered disulfide bridge with the CH3 domain that contains the knob-hole orthogonal mutations described in more detail in section 6.3.14.2 below; an isoallotypic mutation described in more detail in section 6.3.2 above; and an orthogonal mutation described in more detail in section 6.3.14.1 below. In some preferred embodiments, the knob-hole orthogonal mutations combined with the isoallotypic mutations are the following mutational changes: D356E, L358M, T366S, L368A, and Y407V.
[0136] 6.3.11. Domain L (variable region) In the binding molecule, domain L has a variable region domain amino acid sequence. The variable region domain amino acid sequence is an antibody variable region domain amino acid sequence, including VL and VH antibody domain sequences, as described herein. VL and VH sequences are described in more detail above in sections 6.3.1 and 6.3.5, respectively. In a preferred embodiment, domain L has a VH antibody domain sequence.
[0137] 6.3.12. Domain M (Constant Region) In the binding molecule, domain M has a constant region domain amino acid sequence, which are described in more detail in Section 6.3.4 above. In a preferred embodiment, domain M has a constant region sequence that is CH1 from the IgG1 isotype, as discussed in more detail in Section 6.3.4.1. 6.3.13. Pairing of Domains A and F
[0138] In the binding molecule, the Domain A VL amino acid sequence and the Domain F VH amino acid sequence associate to form an antigen binding site (ABS). The A:F antigen binding site (ABS) is capable of specifically binding to an epitope of an antigen. Antigen binding by the ABS is described in further detail below in Section 6.3.13.1.
[0139] In various multivalent embodiments, the ABS formed by domains A and F (A:F) is identical in sequence to one or more other ABSs in the binding molecule, and therefore has the same recognition specificity as one or more other sequence-identical ABSs in the binding molecule.
[0140] In various multivalent embodiments, the A:F ABS is non-identical in sequence to one or more other ABSs in the binding molecule. In certain embodiments, the A:F ABS has a different recognition specificity than one or more other sequence-non-identical ABSs in the binding molecule. In certain embodiments, the A:F ABS recognizes a different antigen than the antigen recognized by at least one other sequence-non-identical ABS in the binding molecule. In certain embodiments, the A:F ABS recognizes a different epitope of an antigen that is also recognized by at least one other sequence-non-identical ABS in the binding molecule. In this embodiment, the ABS formed by domains A and F recognizes an epitope of an antigen, where one or more other ABSs in the binding molecule recognize the same antigen, but not the same epitope.
[0141] 6.3.13.1. Binding of antigens by ABS An ABS and binding molecules that contain such an ABS are said to "recognize" the epitope (or more generally, the antigen) to which the ABS specifically binds, and the epitope (or more generally, the antigen) is said to be the "recognition specificity" or "binding specificity" of the ABS.
[0142] An ABS is said to bind to that particular antigen or epitope with a particular affinity.
[0143] As used herein, "affinity" refers to the strength of the non-covalent intermolecular force interaction between one molecule and another. Affinity, or the strength of the interaction, is determined by the dissociation equilibrium constant (K D ), where K D A lower value indicates a stronger interaction between the molecules. D The affinity can be measured by methods well known in the art, including, but not limited to, biolayer interferometry (e.g., Octet / FORTEBIO®), surface plasmon resonance (SPR) technology (e.g., Biacore®), and cell binding assays. For purposes herein, affinity is the dissociation equilibrium constant measured by biolayer interferometry using Octet / FORTEBIO®.
[0144] "Specific binding," as used herein, refers to the affinity between an ABS and its cognate antigen or epitope, where K D The value is 10 -6 M, 10 -7 M, 10 -8 M, 10 -9 M or 10 -10 Less than M.
[0145] The number of ABS in a binding molecule described herein defines the "valency" of the binding molecule. As outlined in FIG. 2, a binding molecule with a single ABS is "monovalent." A binding molecule with multiple ABS is said to be "multivalent." A multivalent binding molecule with two ABS is "bivalent." A multivalent binding molecule with three ABS is "trivalent." A multivalent binding molecule with four ABS is "tetravalent."
[0146] In various multivalent embodiments, the ABSs all have the same recognition specificity. As outlined in FIG. 2, such binding molecules are "monospecific" "multivalent" binding constructs. In other multivalent embodiments, at least two of the ABSs have different recognition specificities. Such binding molecules are multivalent and "multispecific." In multivalent embodiments, the ABSs collectively have two recognition specificities, the binding molecule is bispecific. In multivalent embodiments, the ABSs collectively have three recognition specificities, the binding molecule is trispecific.
[0147] In multivalent embodiments where the ABS collectively have multiple recognition specificities for different epitopes present on the same antigen, the binding molecule is "multiparatopic." In multivalent embodiments where the ABS collectively recognize two epitopes on the same antigen, the binding molecule is "biparatopic."
[0148] In various multivalent embodiments, the multivalency of the binding molecule enhances the avidity of the binding molecule for a particular target. As described herein, "avidity" refers to the overall strength of interaction between two or more molecules, e.g., multivalent binding molecules for a particular target, where avidity is the cumulative strength of the interaction given by the affinity of multiple ABSs. Avidity can be measured by the same methods used to determine affinity, as described above. In certain embodiments, the avidity of a binding molecule for a particular target is determined by the interaction being a specific binding interaction, where the avidity between two molecules is greater than 10. -6 M, 10 -7 M, 10 -8 M, 10 -9 M or 10 -10 K under M D In certain embodiments, the avidity of a binding molecule for a particular target is determined by a K Dwhere the affinity of one or more of the individual ABSs is sufficient for specific binding to their respective antigen or epitope itself, K D In certain embodiments, avidity is the cumulative strength of interaction provided by the affinity of multiple ABSs to distinct antigens of a common specific target or complex, such as distinct antigens found on individual cells. In certain embodiments, avidity is the cumulative strength of interaction provided by the affinity of multiple ABSs to distinct epitopes of a common individual antigen.
[0149] 6.3.14. Pairing of Domains B and G In the binding molecules described herein, the CH3 amino acid sequence of domain B is associated with the CH3 amino acid sequence of domain G. The CH3 sequence is described in further detail in Section 6.3.2, above.
[0150] In various embodiments, the amino acid sequences of the B and G domains are identical. In some of these embodiments, the sequence is the endogenous CH3 sequence.
[0151] In various embodiments, the amino acid sequences of the B and G domains are distinct and each separately contains an orthogonal modification to the endogenous CH3 sequence, where the B domain interacts with the G domain, and where neither the B domain nor the G domain interacts significantly with the CH3 domain that lacks the orthogonal modification.
[0152] An "orthogonal modification" or synonymously "orthogonal mutation" as described herein is one or more engineered mutations in the amino acid sequence of an antibody domain that increases the affinity of binding of a first domain having an orthogonal modification to a second domain having a complementary orthogonal modification. In certain embodiments, the orthogonal modification decreases the affinity of a domain having an orthogonal modification to a domain lacking the complementary orthogonal modification. In certain embodiments, the orthogonal modification is a mutation in the endogenous antibody domain sequence. In various embodiments, the orthogonal modification is a modification of the N-terminus or C-terminus of the endogenous antibody domain sequence, including, but not limited to, the addition or deletion of amino acids. In certain embodiments, the orthogonal modification includes, but is not limited to, engineered disulfide bridges, knob-in-hole mutations, and charge pair mutations, which are described in more detail below in Sections 6.3.14.1-6.3.14.3. In certain embodiments, the orthogonal modification includes a combination of orthogonal modifications selected from, but not limited to, engineered disulfide bridges, knob-in-hole mutations, and charge pair mutations. In certain embodiments, orthogonal modifications can be combined with amino acid substitutions that reduce immunogenicity, such as isoallotypic mutations, which are described in further detail in Section 6.3.2, above.
[0153] 6.3.14.1. Orthogonally Engineered Disulfide Bridges In various embodiments, the orthogonal modification comprises a mutation that generates an engineered disulfide bridge between the first and second domains. As described herein, an "engineered disulfide bridge" is a mutation that provides two or more domains with a non-endogenous cysteine amino acid so that the two or more domains form a non-natural disulfide bond when they associate. Engineered disulfide bridges are described in more detail in Merchant et al. (Nature Biotech (1998) 16:677-681), which is incorporated herein by reference in its entirety for all teachings. In certain embodiments, the engineered disulfide bridge improves the orthogonal association between certain domains. In certain embodiments, the mutation that generates the engineered disulfide bridge is a K392C mutation in one of the first or second CH3 domains and a D399C in the other CH3 domain. In a preferred embodiment, the mutations generating the engineered disulfide bridges are the S354C mutation in one of the first or second CH3 domains and the Y349C mutation in the other CH3 domain. In another preferred embodiment, the mutations generating the engineered disulfide bridges are the 447C mutations in both the first and second CH3 domains provided by the C-terminal extension of the CH3 domains incorporating the KSC tripeptide sequence.
[0154] 6.3.14.2. Orthogonal Knob-Hole Mutations In various embodiments, the orthogonal modification comprises a knob-hole (synonymously knob-in-hole) mutation. As described herein, a knob-hole mutation is a mutation that changes the conformational features of the surface of a first domain such that the first domain preferentially associates with a second domain having a complementary conformational mutation compared to association with the domain without the complementary conformational mutation. Knob-hole mutations are described in more detail in U.S. Pat. Nos. 5,821,333 and 8,216,805, each of which is incorporated herein in its entirety. In various embodiments, knob-hole mutations are combined with engineered disulfide bridges as described in more detail in Merchant et al. (Nature Biotech (1998) 16:677-681), which is incorporated herein by reference in its entirety. In various embodiments, knob-hole mutations, isoallotypic mutations, and engineered disulfide mutations are combined.
[0155] In certain embodiments, the knob-in-hole mutations are a T366Y mutation in the first domain and a Y407T mutation in the second domain. In certain embodiments, the knob-in-hole mutations are F405A in the first domain and T394W in the second domain. In certain embodiments, the knob-in-hole mutations are a T366Y mutation and F405A in the first domain and a T394W and Y407T in the second domain. In certain embodiments, the knob-in-hole mutations are a T366W mutation in the first domain and a Y407A mutation in the second domain. In certain embodiments, the combined knob-in-hole mutations and engineered disulfide mutations are an S354C and a T366W mutation in the first domain and a Y349C, T366S, L368A and Y407V mutation in the second domain. In a preferred embodiment, the combined knob-in-hole, isoallotypic, and engineered disulfide mutations are S354C and T366W mutations in the first domain, and Y349C, D356E, L358M, T366S, L368A, and Y407V mutations in the second domain.
[0156] Orthogonal Charge Pair Mutations In various embodiments, the orthogonal modification is a charge-pair mutation. As used herein, a charge-pair mutation is a mutation that affects the charge of the amino acid on the surface of a domain, so that the domain preferentially associates with a second domain that has a complementary charge-pair mutation, compared to the association with the domain without the complementary charge-pair mutation. In certain embodiments, the charge-pair mutation improves the orthogonal association between certain domains. Charge-pair mutations are described in more detail in U.S. Patent No. 8,592,562, U.S. Patent No. 9,248,182, and U.S. Patent No. 9,358,286, each of which is incorporated herein by reference for all teachings. In certain embodiments, the charge-pair mutation improves the stability between certain domains. In a preferred embodiment, the charge-pair mutation is a T366K mutation in the first domain, and a L351D mutation in the other domain.
[0157] 6.3.15. Pairing of Domains E and K In various embodiments, the E domain has a CH3 amino acid sequence.
[0158] In various embodiments, the K domain has a CH3 amino acid sequence.
[0159] In various embodiments, the amino acid sequences of the E and K domains are identical and the sequence is an endogenous CH3 sequence.
[0160] In various embodiments, the sequences of the E and K domains are different. In various embodiments, the different sequences each separately contain an orthogonal modification to the endogenous CH3 sequence, where the E domain interacts with the K domain, and neither the E domain nor the K domain significantly interacts with the CH3 domain, which lacks the orthogonal modification. In certain embodiments, the orthogonal modifications include, but are not limited to, engineered disulfide bridges, knob-in-hole mutations, and charge pair mutations, as described in more detail above in Sections 6.3.14.1-6.3.14.3. In certain embodiments, the orthogonal modifications include a combination of orthogonal modifications selected from, but not limited to, engineered disulfide bridges, knob-in-hole mutations, and charge pair mutations. In certain embodiments, the orthogonal modifications can be combined with amino acid substitutions that reduce immunogenicity, such as isoallotypic mutations.
[0161] In various embodiments, the amino acid sequences of the E domain and the K domain are endogenous sequences of two different antibody domains that are selected to have specific interactions that promote specific association between the first polypeptide and the third polypeptide. In various embodiments, the two different amino acid sequences are CH1 and CL sequences. CH1 and CL sequences are described in more detail in sections 6.3.4.1 and 6.3.4.2 above, respectively. The use of CH1 and CL sequences at the C-terminus of heavy chains to promote specific heavy chain association is described in U.S. Patent No. 8,242,247, which is incorporated herein by reference in its entirety for all teachings. In certain embodiments, both CH1 and CL sequences are endogenous sequences. In certain embodiments, the CH1 and CL sequences each separately comprise orthogonal modifications to the endogenous CH1 and CL sequences. In certain embodiments, the orthogonal modification in the endogenous CH1 and CL sequences is an engineered disulfide bridge selected from engineered cysteines at positions 138 in the CH1 sequence and 116 in the CL sequence, 128 in the CH1 sequence and 119 in the CL sequence, or 129 in the CH1 sequence and 210 in the CL sequence, as numbered and discussed in more detail in U.S. Patent Nos. 8,053,562 and 9,527,927, each of which is incorporated by reference in its entirety. In a preferred embodiment, the engineered cysteines are at positions 128 in the CH1 sequence and 118 in the CL kappa sequence, as numbered by the Eu index.
[0162] 6.3.16. Pairing of Domains I and M and Pairing of Domains H and L In various embodiments, Domain I has a CL sequence and Domain M has a CH1 sequence. In various embodiments, Domain H has a VL sequence and Domain L has a VH sequence. In a preferred embodiment, Domain H has a VL amino acid sequence, Domain I has a CL amino acid sequence, Domain L has a VH amino acid sequence, and Domain M has a CH1 amino acid sequence. In another preferred embodiment, Domain H has a VL amino acid sequence, Domain I has a CL amino acid sequence, Domain L has a VH amino acid sequence, Domain M has a CH1 amino acid sequence, and Domain K has a CH3 amino acid sequence.
[0163] In various embodiments, the amino acid sequences of the I domain and the M domain each separately contain an orthogonal modification in the endogenous sequence, where the I domain interacts with the M domain, and neither the I domain nor the M domain interacts significantly with the domain lacking the orthogonal modification. In one set of embodiments, the orthogonal mutation in the I domain is in the CL sequence, and the orthogonal mutation in the M domain is in the CH1 sequence. Orthogonal mutations are described in more detail in Sections 6.3.14.1-6.3.14.3 above. In various embodiments, the orthogonal mutations in the CL sequence and the CH1 sequence are charge pair mutations. In certain embodiments, the charge pair mutation is A141L in the CH1 sequence corresponding to a F118S, F118A or F118V mutation in the CL sequence numbered by the Eu index, or K147D in the CH1 sequence corresponding to a T129R mutation in the CL sequence, as further described in detail in Bonisch et al. (Protein Engineering, Design & Selection, 2017, pp. 1-12), which is incorporated herein by reference for all its teachings. In a preferred set of embodiments, the charge pair mutation is G166D in the CH1 sequence corresponding to a N138K mutation in the CL sequence numbered by the Eu index, or G166K in the CH1 sequence corresponding to a N138D mutation in the CL sequence.
[0164] In various embodiments, orthogonal mutations in the CL and CH1 sequences generate engineered disulfide bridges. In a preferred set of embodiments, the mutation providing a non-intrinsic cysteine amino acid is A141C in the CH1 sequence corresponding to the F118C mutation in the CL sequence, or L128C in the CH1 sequence corresponding to the F118C mutation in the CL sequence, or P171C in the CH1 sequence corresponding to the S162C mutation in the CL sequence, as numbered by the Eu index.
[0165] In various embodiments, the amino acid sequences of the H domain and the L domain each separately contain an orthogonal modification in the endogenous sequence, where the H domain interacts with the L domain, and neither the H domain nor the L domain significantly interacts with the domain lacking the orthogonal modification. In one set of embodiments, the orthogonal mutation in the H domain is in the VL sequence, and the orthogonal mutation in the L domain is in the VH sequence. In a particular embodiment, the orthogonal mutation is a charge pair mutation at the VH / VL interface. In a preferred embodiment, the charge pair mutation at the VH / VL interface is a Q39E in VH and a corresponding Q38K in VL, or a Q39K in VH and a corresponding Q38E in VL, as described in more detail in Igawa et al. (Protein Eng. Des. Sel., 2010, vol. 23, pp. 667-677), which is incorporated herein by reference for all its teachings.
[0166] In certain embodiments, the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, and the interaction between the H and L domains forms a second antigen-binding site specific for a second antigen. In certain embodiments, the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, and the interaction between the H and L domains forms a second antigen-binding site specific for the first antigen.
[0167] 6.3.17. Trivalent binding molecules In another series of embodiments, the binding molecule has three antigen-binding sites and is therefore referred to as "trivalent."
[0168] With reference to Figure 21, in various trivalent embodiments, the binding molecule further comprises a fifth polypeptide chain, (a) the first polypeptide chain further comprises a domain N and a domain O, where the domains are arranged from N-terminus to C-terminus in a NOABDE orientation, domain N has a VL amino acid sequence, and domain O has a CH3 amino acid sequence; (b) the binding molecule further comprises a fifth polypeptide chain, where the fifth polypeptide chain comprises a domain P and a domain Q, where the domains are arranged from N-terminus to C-terminus in a PQ orientation, domain P has a VH amino acid sequence, and domain Q has a CH3 amino acid sequence; (c) the first and fifth polypeptides associate via interactions between the N and P domains and between the O and Q domains to form the binding molecule. As outlined in Figure 2, these trivalent embodiments are referred to as "2x1" trivalent constructs.
[0169] With reference to Figure 26, in a further series of trivalent embodiments, the binding molecule further comprises a sixth polypeptide chain, (a) the third polypeptide chain further comprises a domain R and a domain S, where the domains are arranged from N-terminus to C-terminus in an orientation of RSHIJK, where domain R has a VL amino acid sequence, and domain S has a constant domain amino acid sequence; (b) the binding molecule further comprises a sixth polypeptide chain, where the sixth polypeptide chain comprises a domain T and a domain U, where the domains are arranged from N-terminus to C-terminus in an orientation of TU, where domain T has a VH amino acid sequence, and domain U has a constant domain amino acid sequence; (c) the third and sixth polypeptides associate via interactions between the R and T domains and between the S and U domains to form the binding molecule. As outlined in Figure 2, these trivalent embodiments are referred to as "1x2" trivalent constructs.
[0170] In various embodiments, Domain O is linked to Domain A via a peptide linker. In various embodiments, Domain S is linked to Domain H via a peptide linker. In a preferred embodiment, the peptide linker linking Domain O to Domain A or Domain S to Domain H is the six amino acid GSGSGS peptide sequence, described in more detail in Section 6.3.19.6.
[0171] Trivalent 2 x 1 bispecific construct [2(AA) x 1(B)] With reference to FIG. 21 , in various embodiments, the amino acid sequences of Domain N and Domain A are identical, the amino acid sequence of Domain H is different from the amino acid sequences of Domain N and A, the amino acid sequences of Domain O and Domain B are identical, the amino acid sequence of Domain I is different from the amino acid sequences of Domain O and B, the amino acid sequences of Domain P and Domain F are identical, the amino acid sequence of Domain L is different from the amino acid sequences of Domain P and F, the amino acid sequences of Domain Q and Domain G are identical, and the amino acid sequence of Domain M is different from the amino acid sequences of Domain Q and G; the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, the interaction between the H and L domains forms a second antigen-binding site specific for a second antigen, and Domain N and Domain P form a third antigen-binding site specific for the first antigen.
[0172] Trivalent 2 x 1 bispecific construct [2(AB) x 1(A)] With reference to FIG. 21 , in various embodiments, the amino acid sequences of Domain N and Domain H are identical, the amino acid sequence of Domain A is different from the amino acid sequences of Domain N and H, the amino acid sequences of Domain O and Domain I are identical, the amino acid sequence of Domain B is different from the amino acid sequences of Domain O and I, the amino acid sequences of Domain P and Domain L are identical, the amino acid sequence of Domain F is different from the amino acid sequences of Domain P and L, the amino acid sequences of Domain Q and Domain M are identical, and the amino acid sequence of Domain G is different from the amino acid sequences of Domain Q and M; the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, the interaction between the H and L domains forms a second antigen-binding site specific for a second antigen, and Domain N and Domain P form a third antigen-binding site specific for the second antigen.
[0173] Trivalent 2 x 1 trispecific construct [2(AB) x 1(C)] With reference to FIG. 21 , in various embodiments, the amino acid sequences of Domain N, Domain A, and Domain H are different, the amino acid sequences of Domain O, Domain B, and Domain I are different, the amino acid sequences of Domain P, Domain F, and Domain L are different, and the amino acid sequences of Domain Q, Domain G, and Domain M are different; interaction between the A and F domains forms a first antigen binding site specific for a first antigen, interaction between the H and L domains forms a second antigen binding site specific for a second antigen, and Domain N and Domain P form a third antigen binding site specific for a third antigen.
[0174] In certain embodiments, Domain O has a constant region sequence that is a CL from a kappa light chain, and Domain Q has a constant region sequence that is a CH1 from an IgG1 isotype, as described in more detail in Sections 6.3.4.1 and 6.3.4.2. In a preferred embodiment, Domain O and Domain Q have CH3 sequences such that they specifically associate with each other, as discussed in more detail in Section 6.3.14 above.
[0175] 6.3.17.4. Trivalent 2x1 monospecific constructs With reference to FIG. 21 , in various embodiments, the amino acid sequences of Domain N, Domain A, and Domain H are identical, the amino acid sequences of Domain O and Domain B are identical, the amino acid sequences of Domain P, Domain F, and Domain L are identical, and the amino acid sequences of Domain Q and Domain G are identical; interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, interaction between the H and L domains forms a second antigen-binding site specific for the first antigen, and Domain N and Domain P form a third antigen-binding site specific for the first antigen.
[0176] Referring to FIG. 21 , in another set of embodiments, the amino acid sequences of Domain N, Domain A, and Domain H are identical, the amino acid sequences of Domain O, Domain B, and Domain I are identical, the amino acid sequences of Domain P, Domain F, and Domain L are identical, and the amino acid sequences of Domain Q, Domain G, and Domain M are identical; interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, interaction between the H and L domains forms a second antigen-binding site specific for the first antigen, and Domain N and Domain P form a third antigen-binding site specific for the first antigen.
[0177] Trivalent 1 x 2 bispecific constructs [1(A) x 2(BA)] With reference to FIG. 26, in various embodiments, the amino acid sequences of Domain R and Domain A are identical, the amino acid sequence of Domain H is different from the amino acid sequences of Domain R and A, the amino acid sequences of Domain S and Domain B are identical, the amino acid sequence of Domain I is different from the amino acid sequences of Domain S and B, the amino acid sequences of Domain T and Domain F are identical, the amino acid sequence of Domain L is different from the amino acid sequences of Domain T and F, the amino acid sequences of Domain U and Domain G are identical, and the amino acid sequence of Domain M is different from the amino acid sequences of Domain U and Domain G; the interaction between the A and F domains forms a first antigen binding site specific for a first antigen, the interaction between the H and L domains forms a second antigen binding site specific for a second antigen, and Domain R and Domain T form a third antigen binding site specific for the first antigen.
[0178] Trivalent 1 x 2 bispecific constructs [1(A) x 2(BB)] With reference to FIG. 26, in various embodiments, the amino acid sequences of Domain R and Domain H are identical, the amino acid sequence of Domain A is different from the amino acid sequences of Domain R and H, the amino acid sequences of Domain S and Domain I are identical, the amino acid sequence of Domain B is different from the amino acid sequences of Domain S and I, the amino acid sequences of Domain T and Domain L are identical, the amino acid sequence of Domain F is different from the amino acid sequences of Domain T and L, the amino acid sequences of Domain U and Domain M are identical, and the amino acid sequence of Domain G is different from the amino acid sequences of Domain U and M; the interaction between the A and F domains forms a first antigen binding site specific for a first antigen, the interaction between the H and L domains forms a second antigen binding site specific for a second antigen, and Domain R and Domain T form a third antigen binding site specific for the second antigen.
[0179] Trivalent 1 x 2 trispecific constructs [1(A) x 2(BC)] With reference to FIG. 26, in various embodiments, the amino acid sequences of Domain R, Domain A, and Domain H are different, the amino acid sequences of Domain S, Domain B, and Domain I are different, the amino acid sequences of Domain T, Domain F, and Domain L are different, and the amino acid sequences of Domain U, Domain G, and Domain M are different; the interaction between the A and F domains forms a first antigen binding site specific for a first antigen, the interaction between the H and L domains forms a second antigen binding site specific for a second antigen, and Domain R and Domain T form a third antigen binding site specific for a third antigen.
[0180] In certain embodiments, Domain S has a constant region sequence that is a CL from a kappa light chain, and Domain U has a constant region sequence that is a CH1 from an IgG1 isotype, as discussed in more detail in Sections 6.3.4.1 and 6.3.4.2. In a preferred embodiment, Domain S and Domain U have CH3 sequences such that they specifically associate with each other, as discussed in more detail above in Section 6.3.14.
[0181] 6.3.17.8. Trivalent 1 x 2 monospecific constructs With reference to FIG. 26, in various embodiments, the amino acid sequences of Domain R, Domain A, and Domain H are identical, the amino acid sequences of Domain S and Domain B are identical, the amino acid sequences of Domain T, Domain F, and Domain L are identical, and the amino acid sequences of Domain U and Domain G are identical; interaction between the A and F domains forms a first antigen-binding site specific for the first antigen, interaction between the H and L domains forms a second antigen-binding site specific for the first antigen, and Domain R and Domain T form a third antigen-binding site specific for the first antigen.
[0182] 6.3.18. Tetravalent 2x2 bond molecules In various embodiments, the binding molecule has four antigen-binding sites and is therefore referred to as "tetravalent."
[0183] With reference to FIG. 34, in a further set of embodiments, the binding molecule further comprises a fifth and sixth polypeptide chain, (a) the first polypeptide chain further comprises domain N and domain O, where the domains are arranged from N-terminus to C-terminus in a NOABDE orientation; (b) the third polypeptide chain further comprises domain R and domain S, where the domains are arranged from N-terminus to C-terminus in a RSHIJK orientation; (c) the binding molecule further comprises a fifth and sixth polypeptide chain, where the fifth polypeptide chain comprises domain P and domain Q, where the domains are arranged from N-terminus to C-terminus in a PQ orientation, and the sixth polypeptide chain comprises domain T and domain U, where the domains are arranged from N-terminus to C-terminus in a TU orientation; (d) the first and fifth polypeptides associate via interactions between the N and P domains and between the O and Q domains, and the third and sixth polypeptides associate via interactions between the R and T domains and between the S and U domains to form the binding molecule.
[0184] In various embodiments, Domain O is linked to Domain A via a peptide linker, and Domain S is linked to Domain H via a peptide linker. In a preferred embodiment, the peptide linker linking Domain O to Domain A and Domain S to Domain H is the six amino acid GSGSGS peptide sequence, as described in more detail in Section 6.3.19.6.
[0185] 6.3.18.1. Tetravalent 2x2 Bispecific Constructs Referring to FIG. 34, in a series of tetravalent 2×2 bispecific binding molecules, the amino acid sequences of Domain N and Domain A are identical, the amino acid sequences of Domain H and Domain R are identical, the amino acid sequences of Domain O and Domain B are identical, the amino acid sequences of Domain I and Domain S are identical, the amino acid sequences of Domain P and Domain F are identical, the amino acid sequences of Domain L and Domain T are identical, the amino acid sequences of Domain Q and Domain G are identical, and the amino acid sequences of Domain M and Domain U are identical; the interaction between the A and F domains forms a first antigen binding site specific for a first antigen, Domain N and Domain P forms a second antigen binding site specific for the first antigen, the interaction between the H and L domains forms a third antigen binding site specific for a second antigen, and the interaction between the R and T domains forms a fourth antigen binding site specific for the second antigen.
[0186] Referring to FIG. 34, in another series of tetravalent 2×2 bispecific binding molecules, the amino acid sequences of Domain H and Domain A are identical, the amino acid sequences of Domain N and Domain R are identical, the amino acid sequences of Domain I and Domain B are identical, the amino acid sequences of Domain O and Domain S are identical, the amino acid sequences of Domain L and Domain F are identical, the amino acid sequences of Domain P and Domain T are identical, the amino acid sequences of Domain M and Domain G are identical, and the amino acid sequences of Domain Q and Domain U are identical; the interaction between the A and F domains forms a first antigen binding site specific for a first antigen, Domain N and Domain P forms a second antigen binding site specific for a second antigen, the interaction between the H and L domains forms a third antigen binding site specific for the first antigen, and the interaction between the R and T domains forms a fourth antigen binding site specific for the second antigen.
[0187] 6.3.18.2. Tetravalent 2x2 monospecific constructs With reference to FIG. 34, in various embodiments, the amino acid sequences of Domain N, Domain A, Domain H, and Domain R are identical, the amino acid sequences of Domain O and Domain B are identical, the amino acid sequences of Domain I and Domain S are identical, the amino acid sequences of Domain P, Domain F, Domain L, and Domain T are identical, the amino acid sequences of Domain Q and Domain G are identical, and the amino acid sequences of Domain M and Domain U are identical; an interaction between the A and F domains forms a first antigen-binding site specific for the first antigen, Domain N and Domain P form a second antigen-binding site specific for the first antigen, an interaction between the H and L domains forms a third antigen-binding site specific for the first antigen, and an interaction between the R and T domains forms a fourth antigen-binding site specific for the first antigen.
[0188] Referring to FIG. 34, in another series of tetravalent 2×2 monospecific embodiments, the amino acid sequences of Domain N, Domain A, Domain H, and Domain R are identical, the amino acid sequences of Domain I and Domain B are identical, the amino acid sequences of Domain O and Domain S are identical, the amino acid sequences of Domain P, Domain F, Domain L, and Domain T are identical, the amino acid sequences of Domain M and Domain G are identical, and the amino acid sequences of Domain Q and Domain U are identical; the interaction between the A and F domains forms a first antigen binding site specific for the first antigen, Domain N and Domain P form a second antigen binding site specific for the first antigen, the interaction between the H and L domains forms a third antigen binding site specific for the first antigen, and the interaction between the R and T domains forms a fourth antigen binding site specific for the first antigen.
[0189] 6.3.19. DOMAIN JUNCTIONS 6.3.19.1. Junction Linking the VL and CH3 Domains In various embodiments, the amino acid sequence forming the junction between the C-terminus of the VL domain and the N-terminus of the CH3 domain is an engineered sequence. In certain embodiments, one or more amino acids are deleted or added at the C-terminus of the VL domain. In certain embodiments, the junction connecting the C-terminus of the VL domain and the N-terminus of the CH3 domain is one of the sequences listed in Table 2 in Section 6.12.6 below. In certain embodiments, A111 is deleted at the C-terminus of the VL domain. In certain embodiments, one or more amino acids are deleted or added at the N-terminus of the CH3 domain. In certain embodiments, P343 is deleted at the N-terminus of the CH3 domain. In certain embodiments, P343 and R344 are deleted at the N-terminus of the CH3 domain. In certain embodiments, one or more amino acids are deleted or added at both the C-terminus of the VL domain and the N-terminus of the CH3 domain. In certain embodiments, A111 is deleted at the C-terminus of the VL domain, and P343 is deleted at the N-terminus of the CH3 domain.In a preferred embodiment, A111 and V110 are deleted at the C-terminus of the VL domain.In another preferred embodiment, A111 and V110 are deleted at the C-terminus of the VL domain, and the N-terminus of the CH3 domain has a P343V mutation.
[0190] 6.3.19.2. Junction Linking the VH and CH3 Domains In various embodiments, the amino acid sequence forming the junction between the C-terminus of the VH domain and the N-terminus of the CH3 domain is an engineered sequence. In certain embodiments, one or more amino acids are deleted or added at the C-terminus of the VH domain. In certain embodiments, the junction connecting the C-terminus of the VH domain and the N-terminus of the CH3 domain is one of the sequences listed in Table 3 in Section 6.12.6 below. In certain embodiments, K177 and G118 are deleted at the C-terminus of the VH domain. In certain embodiments, one or more amino acids are deleted or added at the N-terminus of the CH3 domain. In certain embodiments, P343 is deleted at the N-terminus of the CH3 domain. In certain embodiments, P343 and R344 are deleted at the N-terminus of the CH3 domain. In certain embodiments, P343, R344 and E345 are deleted at the N-terminus of the CH3 domain. In certain embodiments, one or more amino acids are deleted or added at both the C-terminus of the VH domain and the N-terminus of the CH3 domain. In a preferred embodiment, T166, K177 and G118 are deleted at the C-terminus of the VH domain.
[0191] 6.3.19.3. Junction (hinge) connecting the C-terminus of CH3 to the N-terminus of CH2 In the binding molecules described herein, the N-terminus of the CH2 domain has a "hinge" region amino acid sequence. As used herein, a hinge region is a sequence of an antibody heavy chain that connects the N-terminal variable domain-constant domain segment of an antibody to the CH2 domain of an antibody. In addition, the hinge region typically provides both flexibility between the N-terminal variable domain-constant domain segment and the CH2 domain, as well as an amino acid sequence motif that forms a disulfide bridge between the heavy chains (e.g., the first and third polypeptide chains). As used herein, the hinge region amino acid sequence is SEQ ID NO:56.
[0192] In various embodiments, the CH3 amino acid sequence is extended at the C-terminus at the junction between the C-terminus of the CH3 domain and the N-terminus of the CH2 domain.In certain embodiments, the CH3 amino acid sequence is extended at the C-terminus at the junction between the C-terminus of the CH3 domain and the hinge region, and then linked to the N-terminus of the CH2 domain.In a preferred embodiment, the CH3 amino acid sequence is extended by inserting a PGK tripeptide sequence, followed by inserting the DKTHT motif of the IgG1 hinge region.
[0193] In certain embodiments, the extension at the C-terminus of the CH3 domain incorporates an amino acid sequence capable of forming a disulfide bond with an orthogonal C-terminal extension of another CH3 domain. In a preferred embodiment, the extension at the C-terminus of the CH3 domain incorporates a KSC tripeptide sequence followed by a DKTHT motif of the IgG1 hinge region, which forms a disulfide bond with an orthogonal C-terminal extension of another CH3 domain incorporating the GEC motif of the kappa light chain.
[0194] 6.3.19.4. Junction (hinge) connecting the C-terminus of CL to the N-terminus of CH2 In various embodiments, the CL amino acid sequence is linked via its C-terminus to the hinge region, which is then linked to the N-terminus of the CH2 domain. The hinge region sequence is described in more detail in Section 6.3.19.3 above. In a preferred embodiment, the hinge region amino acid sequence is SEQ ID NO:56.
[0195] 6.3.19.5. Junction Linking the C-Terminus of CH2 to the Constant Region Domain In various embodiments, the CH2 amino acid sequence is linked to the N-terminus of the constant region domain via its C-terminus. Constant regions are described in more detail in section 6.3.4 above. In a preferred embodiment, the CH2 sequence is linked to the CH3 sequence via its endogenous sequence. In other embodiments, the CH2 sequence is linked to the CH1 or CL sequence. Examples that discuss the linking of the CH2 sequence to the CH1 or CL sequence are described in more detail in U.S. Patent No. 8,242,247, which is incorporated herein in its entirety.
[0196] 6.3.19.6. Junctions Linking Domain O to Domain A or Domain S to Domain H in Trivalent and Tetravalent Molecules In various embodiments, the heavy chains of the antibody (e.g., the first and third polypeptide chains) are extended at their N-terminus to include additional domains that provide additional ABS. With reference to Figures 21, 26 and 34, in certain embodiments, the C-terminus of the constant region domain amino acid sequence of domain O and / or domain S is linked to the N-terminus of the variable region domain amino acid sequence of domain A and / or domain H, respectively. In some preferred embodiments, the constant region domain is a CH3 amino acid sequence and the variable region domain is a VL amino acid sequence. In some preferred embodiments, the constant region domain is a CL amino acid sequence and the variable region domain is a VL amino acid sequence. In certain embodiments, the constant region domain is linked to the variable region domain via a peptide linker. In a preferred embodiment, the peptide linker is the 6 amino acid GSGSGS peptide sequence.
[0197] In various embodiments, the light chain of the antibody (e.g., the second and fourth polypeptide chains) is extended at its N-terminus to include an additional variable domain-constant domain segment of the antibody. In certain embodiments, the constant region domain is a CH1 amino acid sequence and the variable region domain is a VH amino acid sequence.
[0198] 6.4. Specific Bivalent Binding Molecules In a further aspect, bivalent binding molecules are provided.
[0199] With reference to FIG. 3 , in a first set of embodiments, the binding molecule comprises first, second, third and fourth polypeptide chains, where (a) the first polypeptide chain comprises Domain A, Domain B, Domain D and Domain E, where the domains are arranged from N-terminus to C-terminus in an ABDE orientation, and Domain A has a VL amino acid sequence, Domain B has a CH3 amino acid sequence, Domain D has a CH2 amino acid sequence and Domain E has a constant region domain amino acid sequence; (b) the second polypeptide chain comprises Domain F and Domain G, where the domains are arranged from N-terminus to C-terminus in an FG orientation, and Domain F has a VH amino acid sequence and Domain G has a CH3 amino acid sequence; (c) the third polypeptide chain comprises Domain H, Domain I, Domain J and Domain K, where the domains are arranged from N-terminus to C-terminus in an HIJK orientation, and Domain H is a variable region domain. (d) a fourth polypeptide chain comprises domain L and domain M, where the domains are arranged N-terminus to C-terminus in an LM orientation, domain L has a variable region domain amino acid sequence and domain M has a constant region domain amino acid sequence; (e) the first and second polypeptides associate through interactions between the A and F domains and between the B and G domains; (f) the third and fourth polypeptides associate through interactions between the H and L domains and between the I and M domains; (g) the first and third polypeptides associate through interactions between the D and J domains and between the E and K domains to form a binding molecule.
[0200] In a preferred embodiment, domain E has a CH3 amino acid sequence; domain H has a VL amino acid sequence; domain I has a CL amino acid sequence, domain K has a CH3 amino acid sequence; domain L has a VH amino acid sequence; and domain M has a CH1 amino acid sequence.
[0201] In certain embodiments, the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, and the interaction between the H and L domains forms a second antigen-binding site specific for a second antigen, and the binding molecule is a bispecific bivalent binding molecule. In certain embodiments, the interaction between the A and F domains forms a first antigen-binding site specific for a first antigen, and the interaction between the H and L domains forms a second antigen-binding site specific for the first antigen, and the binding molecule is a monospecific bivalent binding molecule.
[0202] 6.4.1. Bivalent, bispecific B-Body "BC1" 3 and 6, in one set of embodiments, the binding molecule has first, second, third and fourth polypeptide chains, where (a) the first polypeptide chain comprises domain A, domain B, domain D and domain E, where the domains are arranged from N-terminus to C-terminus in an ABDE orientation, domain A has a first VL amino acid sequence, domain B has a human IgG1 CH3 amino acid sequence with a T366K mutation and a C-terminal extension incorporating a KSC tripeptide sequence followed by a DKTHT motif of the IgG1 hinge region, domain D has a human IgG1 CH2 amino acid sequence and domain E has a human IgG1 CH3 amino acid sequence with S354C and T366W mutations; (b) the second polypeptide chain comprises domain F and domain G, where the domains are arranged from N-terminus to C-terminus in an FG orientation, domain F has a first VH amino acid sequence and domain G has a human IgG1 CH3 amino acid sequence with a L351D mutation and a C-terminal extension incorporating a GEC amino acid disulfide motif. (c) the third polypeptide chain has Domain H, Domain I, Domain J, and Domain K, where the domains are arranged, from N-terminus to C-terminus, in an HIJK orientation, and Domain H has a second VL amino acid sequence, Domain I has a human CL kappa amino acid sequence, Domain J has a human IgG1 CH2 amino acid sequence, and Domain K has a human IgG1 CH3 amino acid sequence with Y349C, D356E, L358M, T366S, L368A, and Y407V mutations; (d) the fourth polypeptide chain has Domain L and Domain M, where the domains are arranged, from N-terminus to C-terminus, in an LM orientation, and Domain L has a second VH amino acid sequence and Domain M has a human IgG1 having a CH1 amino acid sequence; (e) the first and second polypeptides associate through an interaction between the A domain and the F domain and an interaction between the B domain and the G domain; (f) the third and fourth polypeptides associate through an interaction between the H domain and the L domain and an interaction between the I domain and the M domain;(g) the first and third polypeptides associate through interactions between the D and J domains and between the E and K domains to form a binding molecule; (h) domain A and domain F form a first antigen-binding site specific for a first antigen; (i) domain H and domain L form a second antigen-binding site specific for a second antigen;
[0203] In a preferred embodiment, the first polypeptide chain has the sequence of SEQ ID NO:8, the second polypeptide chain has the sequence of SEQ ID NO:9, the third polypeptide chain has the sequence of SEQ ID NO:10, and the fourth polypeptide chain has the sequence of SEQ ID NO:11.
[0204] 6.4.2. Bivalent, bispecific B-Body "BC6" 3 and 14, in one set of embodiments, the binding molecule has first, second, third and fourth polypeptide chains, where (a) the first polypeptide chain comprises domain A, domain B, domain D and domain E, where the domains are arranged from N-terminus to C-terminus in an ABDE orientation, domain A has a first VL amino acid sequence, domain B has a human IgG1 CH3 amino acid sequence with a KSC tripeptide sequence followed by a C-terminal extension incorporating a DKTHT motif of the IgG1 hinge region, domain D has a human IgG1 CH2 amino acid sequence and domain E has a human IgG1 CH3 amino acid sequence with S354C and T366W mutations; (b) the second polypeptide chain comprises domain F and domain G, where the domains are arranged from N-terminus to C-terminus in an FG orientation, domain F has a first VH amino acid sequence and domain G has a human IgG1 CH3 amino acid sequence with a C-terminal extension incorporating a GEC amino acid disulfide motif. (c) the third polypeptide chain has Domain H, Domain I, Domain J, and Domain K, where the domains are arranged, from N-terminus to C-terminus, in an HIJK orientation, and Domain H has a second VL amino acid sequence, Domain I has a human CL kappa amino acid sequence, Domain J has a human IgG1 CH2 amino acid sequence, and Domain K has a human IgG1 CH3 amino acid sequence with Y349C, D356E, L358M, T366S, L368A, and Y407V mutations; (d) the fourth polypeptide chain has Domain L and Domain M, where the domains are arranged, from N-terminus to C-terminus, in an LM orientation, and Domain L has a second VH amino acid sequence and Domain M has a human IgG1 having a CH1 amino acid sequence; (e) the first and second polypeptides associate through an interaction between the A and F domains and an interaction between the B and G domains; (f) the third and fourth polypeptides associate through an interaction between the H and L domains and an interaction between the I and M domains; (g) the first and third polypeptides associate through an interaction between the D and J domains and an interaction between the E and K domains to form a binding molecule;(h) Domain A and Domain F form a first antigen-binding site specific for a first antigen; (i) Domain H and Domain L form a second antigen-binding site specific for a second antigen;
[0205] 6.4.3. Bivalent, bispecific B-Body "BC28" 3 and 16, in one series of embodiments, the binding molecule has a first, second, third and fourth polypeptide chain, where (a) the first polypeptide chain comprises domain A, domain B, domain D and domain E, where the domains are arranged from N-terminus to C-terminus in an ABDE orientation, domain A has a first VL amino acid sequence, domain B has a human IgG1 CH3 amino acid sequence with a Y349C mutation and a C-terminal extension incorporating a PGK tripeptide sequence followed by a DKTHT motif of the IgG1 hinge region, domain D has a human IgG1 CH2 amino acid sequence and domain E has a human IgG1 CH3 amino acid sequence with S354C and T366W mutations; (b) the second polypeptide chain comprises domain F and domain G, where the domains are arranged from N-terminus to C-terminus in an FG orientation, domain F has a first VH amino acid sequence and domain G has a human IgG1 CH3 amino acid sequence with a S354C mutation and a C-terminal extension incorporating a PGK tripeptide sequence. (c) the third polypeptide chain has Domain H, Domain I, Domain J, and Domain K, where the domains are arranged, from N-terminus to C-terminus, in an HIJK orientation, and Domain H has a second VL amino acid sequence, Domain I has a human CL kappa amino acid sequence, Domain J has a human IgG1 CH2 amino acid sequence, and Domain K has a human IgG1 CH3 amino acid sequence with Y349C, D356E, L358M, T366S, L368A, and Y407V; (d) the fourth polypeptide chain has Domain L and Domain M, where the domains are arranged, from N-terminus to C-terminus, in an LM orientation, and Domain L has a second VH amino acid sequence and Domain M has a human IgG1 having a CH1 amino acid sequence; (e) the first and second polypeptides associate through an interaction between the A and F domains and an interaction between the B and G domains; (f) the third and fourth polypeptides associate through an interaction between the H and L domains and an interaction between the I and M domains; (g) the first and third polypeptides associate through an interaction between the D and J domains and an interaction between the E and K domains to form a binding molecule;(h) Domain A and Domain F form a first antigen-binding site specific for a first antigen; (i) Domain H and Domain L form a second antigen-binding site specific for a second antigen;
[0206] In a preferred embodiment, the first polypeptide chain has the sequence of SEQ ID NO:24, the second polypeptide chain has the sequence of SEQ ID NO:25, the third polypeptide chain has the sequence of SEQ ID NO:10, and the fourth polypeptide chain has the sequence of SEQ ID NO:11.
[0207] 6.4.4. Bivalent, bispecific B-Body "BC44" 3 and 19, in one series of embodiments, the binding molecule has first, second, third and fourth polypeptide chains, where (a) the first polypeptide chain comprises Domain A, Domain B, Domain D and Domain E, where the domains are arranged N-terminus to C-terminus in an ABDE orientation, Domain A has a first VL amino acid sequence, Domain B has a human IgG1 CH3 amino acid sequence with a Y349C mutation, a P343V mutation and a C-terminal extension incorporating a PGK tripeptide sequence followed by a DKTHT motif of the IgG1 hinge region, Domain D has a human IgG1 CH2 amino acid sequence and Domain E has a human IgG1 CH3 amino acid sequence with a S354C mutation and a T366W mutation. (b) the second polypeptide chain has Domain F and Domain G, where the domains are arranged N-terminus to C-terminus in an FG orientation, where Domain F has a first VH amino acid sequence and Domain G has a human IgG1 CH3 amino acid sequence with a S354C mutation and a C-terminal extension incorporating a PGK tripeptide sequence; (c) the third polypeptide chain has Domain H, Domain I, Domain J, and Domain K, where the domains are arranged N-terminus to C-terminus in an HIJK orientation, where Domain H has a second VL amino acid sequence, Domain I has a human CL kappa amino acid sequence, Domain J has a human IgG1 CH2 amino acid sequence, and Domain K has a human IgG1 CH2 amino acid sequence with Y349C, T366S, L368A, and Y407V. (d) a fourth polypeptide chain has a domain L and a domain M, where the domains are arranged N-terminus to C-terminus in an LM orientation, wherein domain L has a second VH amino acid sequence and domain M has a human IgG1 amino acid sequence; (e) the first and second polypeptides associate through interactions between the A and F domains and between the B and G domains; (f) the third and fourth polypeptides associate through interactions between the H and L domains and between the I and M domains; (g) the first and third polypeptides associate through interactions between the D and J domains and between the E and K domains to form a binding molecule;(h) Domain A and Domain F form a first antigen-binding site specific for a first antigen; (i) Domain H and Domain L form a second antigen-binding site specific for a second antigen;
[0208] In a preferred embodiment, the first polypeptide chain has the sequence of SEQ ID NO: 32, the second polypeptide chain has the sequence of SEQ ID NO: 25, the third polypeptide chain has the sequence of SEQ ID NO: 10, and the fourth polypeptide chain has the sequence of SEQ ID NO: 11.
[0209] 6.5. Specific Trivalent Binding Molecules 6.5.1. Trivalent 1×2 bispecific B-Body "BC28-1×2" With reference to Section 6.4.3 and Figure 26, in one set of embodiments the binding molecule further comprises a sixth polypeptide chain, wherein (a) the third polypeptide chain further comprises Domain R and Domain S, wherein the domains are arranged N-terminus to C-terminus in an orientation of RSHIJK, where Domain R has a first VL amino acid sequence and Domain S has a human IgG1 CH3 amino acid sequence with a Y349C mutation and a C-terminal extension incorporating a PGK tripeptide sequence followed by a GSGSGS linker peptide linking Domain S to Domain H; (b) the binding molecule further comprises a sixth polypeptide chain, wherein the sixth polypeptide chain comprises Domain T and Domain U, wherein the domains are arranged N-terminus to C-terminus in an orientation of TU, where Domain T has the first VH amino acid sequence and Domain U has a S354C mutation and a C-terminal extension incorporating a PGK tripeptide sequence. (c) the third and sixth polypeptides associate via interactions between the R and T domains and between the S and U domains to form the binding molecule; and (d) the R and T domains form a third antigen-binding site specific for the first antigen.
[0210] In a preferred embodiment, the first polypeptide chain has the sequence of SEQ ID NO:24, the second polypeptide chain has the sequence of SEQ ID NO:25, the third polypeptide chain has the sequence of SEQ ID NO:37, the fourth polypeptide chain has the sequence of SEQ ID NO:11, and the sixth polypeptide chain has the sequence of SEQ ID NO:25.
[0211] 6.5.2. Trivalent 1x2 trispecific B-Body "BC28-1x1x1a" With reference to Section 6.4.3 and Figures 26 and 30, in one set of embodiments the binding molecule further comprises a sixth polypeptide chain, wherein (a) the third polypeptide chain further comprises Domain R and Domain S, wherein the domains are arranged N-terminus to C-terminus in a RSHIJK orientation, where Domain R has a third VL amino acid sequence, and Domain S has a human IgG1 CH3 amino acid sequence with a T366K mutation and a C-terminal extension incorporating a KSC tripeptide sequence followed by a GSGSGS linker peptide linking Domain S to Domain H; (b) the binding molecule further comprises a sixth polypeptide chain, wherein the sixth polypeptide chain further comprises Domain T and Domain U, wherein the domains are arranged N-terminus to C-terminus in a TU orientation, where Domain T has a third VH amino acid sequence, and Domain U has a L351D mutation and a C-terminal extension incorporating a GEC amino acid disulfide motif. (c) the third and sixth polypeptides associate via interactions between the R and T domains and between the S and U domains to form a binding molecule; and (d) the R and T domains form a third antigen-binding site specific for a third antigen.
[0212] In a preferred embodiment, the first polypeptide chain has the sequence of SEQ ID NO:24, the second polypeptide chain has the sequence of SEQ ID NO:25, the third polypeptide chain has the sequence of SEQ ID NO:45, the fourth polypeptide chain has the sequence of SEQ ID NO:11, and the sixth polypeptide chain has the sequence of SEQ ID NO:53.
[0213] 6.6. Antigen specificity The antigen-binding sites of the binding molecules described herein may be selected to specifically bind to a wide variety of molecular targets. For example, the antigen binding site(s) may be any of the following: E-Cad, CLDN7, FGFR2b, N-Cad, Cad-11, FGFR2c, ERBB2, ERBB3, FGFR1, FOLR1, IGF-Ira, GLP1R, PDGFRa, PDGFRb, EPHB6, ABCG2, CXCR4, CXCR7, integrin-avb3, SPARC, VCAM, ICAM, annexin, ROR1, ROR2, TNFα, CD137, angiopoietin 2, angiopoietin 3, BAFF, beta amyloid, C5, CA-125, CD147, CD125, CD147, CD152, CD19, CD20, CD22, CD23, CD24, CD25, CD274, CD28, CD3, CD30, CD33, CD37, CD4, CD 40, CD44, CD44v4, CD44v6, CD44v7, CD50, CD51, CD52, CEA, CSF1R, CTLA-2, DLL4, EGFR, EPCAM, HER3, GD2 ganglioside, GDF-8, Her2 / neu, CD2221, IL-17A, IL-12, IL-23, IL-13, IL-6, IL-23, integrin, CD11a, MUC1, Notch, TAG-72, TGFβ, TRAIL-R2, VEGF-A, VEGFR-1, VEGFR2, VEGFc, hematopoietin (four-helix bundle) (EPO (erythropoietin), IL-2 (T-cell growth factor), IL-3 (multi-colony CSF), IL-4 (BCGF-1, BSF-1), IL-5 (BCGF-2), IL-6 Such as IL-4 (IFN-β2, BSF-2, BCDF), IL-7, IL-8, IL-9, IL-11, IL-13 (P600), G-CSF, IL-15 (T-cell growth factor), GM-CSF (granulocyte-macrophage colony-stimulating factor), OSM (OM, oncostatin M), and LIF (leukemia inhibitory factor); interferons (such as IFN-γ, IFN-α, and IFN-β); and immunoglobulin superfamily (B7.1 (CD80) and B7.2 (B70, CD86, etc.); TNF family (TNF-α (cachectin), TNF-β (lymphotoxin, LT, LT-α), LT-β, Fas, CD27, CD30, and 4-1BBL, etc.); as well as those not classified in a particular family (TGF-β, IL 1α, IL-1β, IL-1 RA, IL-10 (cytokine synthesis inhibitor F), IL-12 (NK cell stimulator), MIF, IL-16, IL-17 (mCTLA-8), and / or IL-18 (IGIF, interferon-γ inducer), etc.); in embodiments relating to bispecific antibodies, the antibody may bind, for example, to two of these targets. Additionally, the Fc portion of the heavy chain of the antibody may be used to target Fc receptor expressing cells, such as the use of the Fc portion of an IgE antibody to target mast cells and basophils.
[0214] The TNF receptor family, e.g., TNFR1 (also known as CD120a and TNFRSF1A), TNFR2 (also known as CD120b and TNFRSF1B), TNFRSF3 (also known as LTβR), TNFRSF4 (also known as OX40 and CD134), TNFRSF5 (also known as CD40), TNFRSF6 (also known as FAS and CD95), TNFRSF6B (also known as DCR3), TNFRSF7 (also known as CD27), T NFRSF8 (also known as CD30), TNFRSF9 (also known as 4-1BB), TNFRSF10A (also known as TRAILR1, DR4, and CD26), TNFRSF10B (also known as TRAILR2, DR5, and CD262), TNFRSF10C (also known as TRAILR3, DCR1, CD263), TNFRSF10D (also known as TRAILR4, DCR2, and CD264), TNFRSF11A (also known as RANK and CD265), also known as CD266), TNFRSF13B (also known as TACI and CD267), TNFRSF13C (also known as BAFFR, BR3, and CD268), TNFRSF14 (also known as HVEM and CD270), TNFRSF16 (also known as NGFR, p75NTR, and CD271), or TNFRSF17 (also known as BCMA and The antigen binding site(s) may be selected that specifically binds to the TNF receptor family, including, but not limited to, TNFRSF18 (also known as GITR and CD357), TNFRSF19 (also known as TROY, TAJ, and TRADE), TNFRSF21 (also known as CD358), TNFRSF25 (also known as Apo-3, TRAMP, LARD, or WS-1), EDA2R (also known as XEDAR).
[0215] The antigen binding site(s) may be selected that specifically binds to an immune-oncology target, for example, a checkpoint inhibitor target such as, but not limited to, PD1, PDL1, CTLA-4, PDL2, B7-H3, B7-H4, BTLA, TIM3, GAL9, LAG3, VISTA, KIR, 2B4, BY55, and CGEN-15049.
[0216] In one set of embodiments, an antigen binding site(s) may be selected that specifically targets tumor-associated cells. In various embodiments, the antigen binding site(s) specifically targets tumor-associated immune cells. In certain embodiments, the antigen binding site(s) specifically targets tumor-associated regulatory T cells (Tregs). In certain embodiments, the binding molecule has an antigen binding site specific for an antigen selected from one or more of CD25, OX40, CTLA-4, and NRP1, such that the binding molecule specifically targets tumor-associated regulatory T cells. In certain embodiments, the binding molecule has an antigen binding site that specifically binds to CD25 and OX40, CD25 and CTLA-4, CD25 and NRP1, OX40 and CTLA-4, OX40 and NRP1, or CTLA-4 and NRP1, such that the binding molecule specifically targets tumor-associated regulatory T cells. In preferred embodiments, the bispecific bivalent binding molecule has antigen binding sites that specifically bind to CD25 and OX40, CD25 and CTLA-4, CD25 and NRP1, OX40 and CTLA-4, OX40 and NRP1, or CTLA-4 and NRP1, such that the binding molecule specifically targets tumor-associated regulatory T cells. In certain embodiments, specific targeting of tumor-associated regulatory T cells leads to depletion (e.g., killing) of the regulatory T cells. In preferred embodiments, depletion of regulatory T cells is mediated by antibody-drug conjugate (ADC) modifications, such as antibodies conjugated to toxins, as discussed in more detail below in Section 6.7.1.
[0217] In one series of embodiments, the binding molecule has an antigen binding site selected from one or more of CD3, ROR1 and ROR2. In a particular embodiment, the bispecific bivalent has an antigen binding site that specifically binds to CD3 and ROR1. In a particular embodiment, the bispecific bivalent has an antigen binding site that specifically binds to CD3 and ROR2. In a particular embodiment, the trispecific trivalent has an antigen binding site that specifically binds to CD3, ROR1 and ROR2.
[0218] 6.7. Further Modifications In a further set of embodiments, the binding molecules have additional modifications.
[0219] 6.7.1. Binding Molecule-Drug Conjugates In various embodiments, the binding molecule is conjugated to a therapeutic agent (i.e., a drug) to form a binding molecule-drug conjugate. Therapeutic agents include, but are not limited to, chemotherapeutic agents, imaging agents (e.g., radioisotopes), immunomodulatory agents (e.g., cytokines, chemokines, or checkpoint inhibitors), and toxins (e.g., cytotoxic agents). In certain embodiments, the therapeutic agent is linked to the binding molecule via a linker peptide, as discussed in more detail in Section 6.7.3 below. Methods for preparing antibody-drug conjugates (ADCs) that may be adapted for conjugating drugs to the binding molecules disclosed herein are described, for example, in U.S. Pat. No. 8,624,003 (the Pott method), U.S. Pat. No. 8,163,888 (the one-step method), U.S. Pat. No. 5,208,020 (the two-step method), U.S. Pat. No. 8,337,856, U.S. Pat. No. 5,773,001, U.S. Pat. No. 7,829,531, U.S. Pat. No. 5,208,020, U.S. Pat. No. 7,745,394, WO2017 / 136623, WO2017 / 015502, WO2017 / 015496, WO2017 / 015495, WO2004 / 010957, each of which is incorporated by reference in its entirety for all its teachings. , WO2005 / 077090, WO2005 / 082023, WO2006 / 065533, WO2007 / 030642, WO2007 / 103288, WO2013 / 17 3337, WO2015 / 057699, WO2015 / 095755, WO2015 / 123679, WO2015 / 157286, WO2017 / 165851, WO200 9 / 073445, WO2010 / 068759, WO2010 / 138719, WO2012 / 171020, WO2014 / 008375, WO2014 / 093394, W O2014 / 093640, WO2014 / 160360, WO2015 / 054659, WO2015 / 195925, WO2017 / 160754, Storz(MAbs., 2015 Nov-Dec;7(6):989-1009), Lambert et al. (Adv Ther, 2017 34:1015), Diamantis et al. (British Journal of Cancer, 2016 114:362-367), Carrico et al. (Nat Chem Biol, 2007 3:321-2), We et al. (Proc Natl Acad Sci USA, 2009 106:3000-5), Rabuka et al. (Curr Opin Chem Biol., 2011 14:790-6), Hudak et al. (Angew Chem Int Ed Engl., 2012 4161-5), Rabuka et al. (Nat Protoc., 2012, 7:1052-67), Agarwal et al. (Proc Natl Acad Sci USA., 2013, 110:46-51), Agarwal et al. (Bioconjugate Chem., 2013, 24:846-851), Barfield et al. (Drug Dev. and D., 2014, 14:34-41), Drake et al. (Bioconjugate Chem., 2014, 25:1331-41), Liang et al. (J Am Chem Soc., 2014, 136:10850-3), Drake et al. (Curr Opin Chem Biol., 2015, 28:174-80) and York et al. (BMC Biotechnology, 2016, Vol. 16 (No. 1): p. 23).
[0220] 6.7.2. Further Bond Substructures In various embodiments, the binding molecule has a modification that includes one or more additional binding moieties. In certain embodiments, the binding moieties are antibody fragments or antibody formats, including, but not limited to, full-length antibodies, Fab fragments, Fvs, scFvs, tandem scFvs, diabodies, sc diabodies, DARTs, tandAbs, minibodies, camelid VHHs, and other antibody fragments or formats known to those skilled in the art. Exemplary antibody and antibody fragment formats are described in detail in Brinkmann et al. (MABS, 2017, Vol. 9, No. 2, pp. 182-212), which is incorporated herein by reference for all its teachings.
[0221] In certain embodiments, one or more additional binding moieties are linked to the C-terminus of the first or third polypeptide chain. In certain embodiments, one or more additional binding moieties are linked to the C-terminus of both the first and third polypeptide chains. In certain embodiments, one or more additional binding moieties are linked to the C-terminus of both the first and third polypeptide chains. In certain embodiments, individual portions of one or more additional binding moieties are linked separately to the C-terminus of the first and third polypeptide chains to form a functional binding moiety.
[0222] In certain embodiments, one or more additional binding moieties are linked to the N-terminus of any of the polypeptide chains (e.g., the first, second, third, fourth, fifth, or sixth polypeptide chain). In certain embodiments, individual portions of the additional binding moieties are separately linked to the N-terminus of different polypeptide chains to form a functional binding moiety.
[0223] In certain embodiments, one or more additional binding moieties are specific for different antigens or epitopes of the ABS in the binding molecule. In certain embodiments, one or more additional binding moieties are specific for the same antigen or epitope of the ABS in the binding molecule. In certain embodiments, where the modification is two or more additional binding moieties, the additional binding moieties are specific for the same antigen or epitope. In certain embodiments, where the modification is two or more additional binding moieties, the additional binding moieties are specific for different antigens or epitopes.
[0224] In certain embodiments, the one or more additional binding moieties are linked to the binding molecule using in vitro methods, including, but not limited to, reactive chemistry and affinity tag systems, as discussed in more detail below in Section 6.7.3. In certain embodiments, the one or more additional binding moieties are linked to the binding molecule via Fc-mediated binding (e.g., Protein A / G). In certain embodiments, the one or more additional binding moieties are linked to the binding molecule using recombinant DNA techniques, such as, for example, encoding the nucleotide sequences of the fusion products of the binding molecule and the additional binding moieties in the same expression vector (e.g., plasmid).
[0225] 6.7.3. Functional / Reactive Groups In various embodiments, the binding molecules have modifications that include functional or chemically reactive groups that can be used in downstream processes, such as, for example, attachment of additional moieties (e.g., drug conjugates and additional binding moieties, discussed in more detail above in Sections 6.7.1 and 6.7.2) and downstream purification processes.
[0226] In certain embodiments, the modification is a chemically reactive group, including, but not limited to, reactive thiols (e.g., maleimide-based reactive groups), reactive amines (e.g., N-hydroxysuccinimide-based reactive groups), "click chemistry" groups (e.g., reactive alkyne groups), and aldehydes bearing formylglycine (FGly). In certain embodiments, the modification is a functional group, including, but not limited to, affinity peptide sequences (e.g., HA, HIS, FLAG, GST, MBP, and Strep systems, etc.). In certain embodiments, the functional group or chemically reactive group has a cleavable peptide sequence. In certain embodiments, the cleavable peptide is cleaved by means including, but not limited to, photocleavage, chemical cleavage, protease cleavage, reducing conditions, and pH conditions. In certain embodiments, the protease cleavage is performed by an intracellular protease. In certain embodiments, the protease cleavage is performed by an extracellular or membrane-associated protease. ADC therapy employing protease cleavage is described in more detail in Choi et al. (Theranostics, 2012;2(2):156-178), which is incorporated by reference in its entirety for all that it teaches.
[0227] Pharmaceutical Compositions In another aspect, there is provided a pharmaceutical composition comprising a binding molecule as described herein and a pharma- ceutically acceptable carrier or diluent. In typical embodiments, the pharmaceutical composition is sterile.
[0228] In various embodiments, the pharmaceutical composition comprises the binding molecule at a concentration of 0.1 mg / ml to 100 mg / ml. In certain embodiments, the pharmaceutical composition comprises the binding molecule at a concentration of 0.5 mg / ml, 1 mg / ml, 1.5 mg / ml, 2 mg / ml, 2.5 mg / ml, 5 mg / ml, 7.5 mg / ml or 10 mg / ml. In some embodiments, the pharmaceutical composition comprises the binding molecule at a concentration of more than 10 mg / ml. In certain embodiments, the binding molecule is present at a concentration of 20 mg / ml, 25 mg / ml, 30 mg / ml, 35 mg / ml, 40 mg / ml, 45 mg / ml, or even 50 mg / ml or higher. In certain embodiments, the binding molecule is present at a concentration of more than 50 mg / ml.
[0229] In various embodiments, pharmaceutical compositions are described in more detail in U.S. Pat. Nos. 8,961,964, 8,945,865, 8,420,081, 6,685,940, 6,171,586, 8,821,865, 9,216,219, U.S. Patent Application No. 10 / 813,483, WO2014 / 066468, WO2011 / 104381, and WO2016 / 180941, each of which is incorporated herein in its entirety.
[0230] 6.9. Manufacturing method The binding molecules described herein can be easily produced by expression using standard cell-free translation, transient transfection and stable transfection approaches currently used for antibody production. In certain embodiments, Expi293 cells (ThermoFisher) can be used for the production of binding molecules using protocols and reagents from ThermoFisher, such as ExpiFectamine, or other reagents known to those skilled in the art, such as polyethyleneimine, as described in detail in Fang et al. (Biological Procedures Online, 2017, vol. 19:11), which is incorporated herein by reference for all teachings.
[0231] As further described in the Examples below, the expressed protein can be readily purified using a CH1 affinity resin, such as CaptureSelect CH1 resin, and protocols provided by ThermoFisher. Further purification may be achieved using ion exchange chromatography as routinely used in the art.
[0232] 6.10. Treatment Method In another aspect, a method of treatment is provided that comprises administering to a patient a binding molecule described herein in an amount effective to treat the patient.
[0233] In some embodiments, the antibodies of the present disclosure may be used to treat cancer, which may be of the bladder, blood, bone, bone marrow, brain, breast, colon, esophageal, gastrointestinal, gum, head, kidney, liver, lung, nasopharyngeal, neck, ovary, prostate, skin, stomach, testis, tongue, or uterus. In some embodiments, the cancer is selected from the group consisting of neoplasms, malignancies, carcinomas, undifferentiated carcinomas, giant cell spindle cell carcinomas, small cell carcinomas, papillary carcinomas, squamous cell carcinomas, lymphomas, basal cell carcinomas, pilomatrix carcinomas, transitional cell carcinomas, papillary transitional cell carcinomas, adenocarcinomas, malignant gastrinomas, cholangiocarcinomas, hepatocellular carcinomas, mixed hepatocellular and cholangiocarcinomas, trabecular adenocarcinomas, adenoid cystic carcinomas, adenocarcinomas of adenomatous polyps, adenomatous familial polyposis coli, solid tumors, carcinoid malignancies, bronchiolo-alveolar adenocarcinomas, papillary adenocarcinomas, chromophobe carcinomas, acidophil carcinomas, acidophil adenocarcinomas, basophilomas, clear cell adenocarcinomas, granular cell carcinomas, follicular adenocarcinomas, papillary and follicular adenocarcinomas, nonencapsulating sclerosing carcinomas, carcinoma);adrenal cortical carcinoma;endometrial carcinoma;skin adnexal carcinoma;apocrine gland carcinoma;sebaceous gland carcinoma;earwax gland carcinoma;mucoepidermoid carcinoma;cystadenocarcinoma;papillary cystadenocarcinoma;papillary serous cystadenocarcinoma;mucinous cystadenocarcinoma;mucinous adenocarcinoma;signet ring cell carcinoma;duct carcinoma invasion;medullary carcinoma;lobular carcinoma;inflammatory tumor;Paget's disease of the breast;acinic cell carcinoma;adenosquamous cell carcinoma;squamous metaplasia-associated adenocarcinoma;malignant thymoma;malignant ovarian stromal tumor;malignant theca cell tumor;malignant granular cell tumor;malignant androgen tumor;Sertoli cell carcinoma;malignant Leydig cell tumor;malignant lipid cell tumor;malignant paraganglioma;malignant extramammary paraganglioma;pheochromocytoma;glomerular sarcoma;malignant melanoma;amelanotic melanoma;superficial spreading melanoma;malignant melanoma of giant pigmented nevus (malig melanoma);epithelioid cell melanoma;malignant blue nevus;sarcoma;fibrosarcoma;malignant fibrous histiocytoma;myxosarcoma;liposarcoma;leiomyosarcoma;rhabdomyosarcoma;embryonal rhabdomyosarcoma;alveolar rhabdomyosarcoma;stromal sarcoma;malignant mixed tumor;Müllerian mixed tumor;nephroblast;hepatoblastoma;carcinosarcoma;malignant mesenchymoma;malignant Brenner tumor;malignant phyllodes tumor;synovial sarcoma;malignant mesothelioma;dysgerminoma;embryonal carcinoma;malignant teratoma;malignant ovarian thyroid tumor;choriocarcinoma;malignant mesonephroma;angiosarcoma;malignant hemangioendothelioma;Kaposi's sarcoma;malignant hemangiopericytoma;lymphangiosarcoma;osteosarcoma;parosteal osteosarcoma;chondrosarcoma;malignant chondroblastoma;Mesenchymal chondrosarcoma;Giant cell tumor of bone;Ewing's sarcoma;Malignant odontogenic tumor;Ameloblastic odontosarcoma;Malignant ameloblastoma;Ameloblastic fibrosarcoma;Malignant pinealoma;Chordoma;Malignant glioma;Ependymoma;Astrocytoma;Protoplasmic astrocytoma;Fibrillary astrocytoma;Astroblastoma;Glioblastoma;Oligodendroglioblastoma;Primitive neuroectodermal;Cerebellar sarcoma;Ganglion;Neuroblastoma;Retinoblastoma;Olfactory neurogenic tumor;Malignant meningioma;Neurofibrosarcoma;Malignant schwannoma;Malignant granular cell tumor;Malignant lymphoma;Hodgkin's disease ;Hodgkin's;lateral granuloma;malignant small lymphocytic lymphoma;malignant diffuse large cell lymphoma;malignant follicular lymphoma;mycosis fungoides;other specified non-Hodgkin's lymphoma;malignant histiocytosis;multiple myeloma;mast cell sarcoma;immunoproliferative small intestinal disease;leukemia;lymphocytic leukemia;plasma cell leukemia;erythroleukemia;lymphoblastic cell leukemia;myeloid leukemia;basophilic leukemia;eosinophilic leukemia;monocytic leukemia;mast cell leukemia;megakaryoblastic leukemia;granulocytic sarcoma;and hairy cell leukemia. ;
[0234] The antibodies of the present disclosure can be administered to a subject per se or in the form of a pharmaceutical composition, for example, for the treatment of cancer, autoimmunity, transplant rejection, immune responses after trauma, graft-versus-host disease, ischemia, stroke, and infectious diseases, for example, by targeting viral antigens such as gp120 of HIV.
[0235] 6.11. Further Aspects and Embodiments Other aspects and embodiments are set forth in the following numbered sections: 1. A domain-swapped antibody comprising a heavy chain (HC) composed of VH-CH3 and a light chain (LC) comprising VL-CH3-CH2-CH3, wherein the VH-CH3 of the HC dimerizes with the VL-CH3 of the LC to form a first domain-swapped LC / HC dimer comprising a CH3LC / CH3HC domain pair. 2. The antibody of item 1, wherein the light chain is further defined as VL-CH3(knob)-CH2-CH3 and the heavy chain is further defined as VH-CH3(hole). 3. The antibody of item 1, wherein the light chain is further defined as VL-CH3(hole)-CH2-CH3 and the heavy chain is further defined as VH-CH3(knob). 4. The antibody according to any of items 1 to 3, wherein the light chain further comprises a second LC / HC dimer, wherein the light chain is further defined as VL-CH3-CH2-CH3-VX1-CHX, and the light chain is further dimerized with VX2-CHX2. 5. The antibody of item 4, wherein VX1 is a second VH region and VX2 is a second VL region. 6. The antibody of item 4, wherein VX1 is a second VL region and VX2 is a second VH region. 7. The antibody according to any one of items 4 to 6, wherein CHX and CHX2 form a CH3 / CH3 dimer. 8. The antibody according to any of items 4 to 6, wherein CHX and CHX2 form a CH1 / CL dimer. 9. The antibody of item 4, wherein the light chain is further defined as VL-CH3-CH2-CH3-VH1-CH1, and VX2-CHX2 is VL1-CL. 10. The antibody according to any of items 1 to 3, wherein the antibody comprises two light chains, wherein the first light chain is further defined as VL-CH3-CH2-CH3-VH1-CH1 and the second light chain is further defined as VL-CH3-CH2-CH3-VL1-CL, wherein VH1-CH1 is dimerized with VL1-CL. 11. The antibody according to any of items 1 to 3, wherein the antibody comprises two light chains, wherein the first light chain is further defined as VL-CH3-CH2-CH3-VH1-CH3 and the second light chain is further defined as VL-CH3-CH2-CH3-VL1-CH3, and VH1-CH3 is dimerized with VL1-CH3. 12. The antibody of item 11, wherein VL1-CH3 is further defined as VL1-CH3(knob) and VH1-CH3 is further defined as VH1-CH3(hole). 13. The antibody according to any of items 1 to 3, wherein the heavy chain further comprises a second LC / HC dimer, wherein the heavy chain is further defined as VX1-CHX-VH-CH3, and wherein the heavy chain is further dimerized with VX2-CHX2. 14. The antibody of item 13, wherein VX1 is a second VH region and VX2 is a second VL region. 15. The antibody of item 13, wherein VX1 is a second VL region and VX2 is a second VH region. 16. The antibody according to any of items 13 to 15, wherein CHX and CHX2 form a CH3 / CH3 dimer. 17. The antibody according to any of items 13 to 15, wherein CHX and CHX2 form a CH1 / CL dimer. 18. The antibody of item 4, wherein the heavy chain is further defined as VH1-CH1-VH-CH3 and VX2-CHX2 is VL1-CL. 19. The antibody according to any of items 4 to 18, wherein the first LC / HC dimer comprises a first binding site that recognizes a first epitope and the second LC / HC dimer comprises a second binding site that recognizes a second epitope that is different from the first epitope or is derived from a different antigen, and wherein either the first LC / HC dimer or the second LC / HC dimer is domain-swapped. 20. The antibody according to item 19, which is a bispecific antibody comprising two first LC / HC dimers and a single second LC / HC dimer. 21. The antibody according to item 19, which is a bispecific antibody comprising two first LC / HC dimers and two second LC / HC dimers. 22. The antibody according to any of items 1 to 21, wherein the HC or LC comprises a mutation as shown in Table A below. [Table A] 23. The antibody according to any one of items 1 to 22, which is an IgG antibody. 24. The antibody according to any of items 1 to 23, wherein LC consists of VL-CH3-CH2-CH3 and optionally further comprises one or more linker / junction or hinge regions. 25. The antibody according to any of items 1 to 24, wherein the first LC / HC dimer is characterized by a CH3(knob) / CH3(hole) dimer, a CH3Hc / CH3Hc dimer, or comprises an engineered CH3 domain capable of generating a cognate pair of CH3LC / CH3HC domains; and the light chain is further dimerized with an Fc chain comprising CH2-CH3, thereby forming an Fc region. 26. The antibody according to any of items 1 to 25, wherein the light chain is further dimerized with an Fc chain comprising a CH2-CH3, thereby forming an Fc region; the Fc region is characterized by a CH3(knob) / CH3(hole) dimer, a CH3Hc / CH3Hc dimer, or comprises an engineered CH3 domain that can give rise to a cognate pair of CH3LC / CH3HC domains. 27. The engineered CH3 domain comprises the amino acid sequence of a CH3 of human IgG1 or a functional variant thereof having at least 80% sequence identity to the CH3 of human IgG1, wherein the engineered CH3 domain comprises one of the following: a) one or more knob or hole mutations, preferably any of the knob or hole mutations T366Y / Y407T, F405A / T394'W, T366Y:F405A / T394'W:Y407T, T366W / Y407'A and S354C:T366W / Y349'C:T366'S:L368'A:Y407V; b) a cysteine residue that is covalently linked to a cysteine residue of the other cognate CH3 domain, thereby introducing an interdomain disulfide bridge, preferably linking the C-termini of both CH3 domains; c) a SEED CH3 heterodimer composed of alternating segments of human IgA and IgG CH3 sequences; and / or d) one or more mutations whose repulsive charges prevent heterodimer formation, preferably one or more mutations that are either K409D / D399, K409D / D399'R, K409E / D399, K409E / D399'R, K409D:K392D / D399'K:E356'K or K409D:K392D:K370D / D399'K:E356'K:E357'K; and / or e) one or more mutations selected for heterodimer formation and / or thermostability, preferably one or more mutations which are either T350V:L351Y:F405A:Y407V / T350V:T366L:K392L:T394W, T350V:L351Y:F405A:Y407V / T350V:T366L:K392M:T394W, L351Y:F405A:Y407V / T366L:K392M:T394W, F405A:Y407V / T366L:K392M:T394W, or F405A:Y407V / T366L:T394W 27. The antibody according to any of items 25 to 26, comprising one or more of the following, wherein the numbering is according to the Eu index: 28. The junction between either the VH or VL domain and the CH3 domain comprises an amino acid sequence, wherein the amino acid sequence is a) at least part of the junction between the CH2 and CH3 domains of a human IgG antibody; and / or b) at least part of the junction between the VL and CL domains of a human IgG antibody; and / or c) at least part of the junction between the VH domain and the CH1 domain of a human IgG antibody; and / or d) The antibody according to any one of items 1 to 25, which is an artificial linked sequence having a length of 5 to 20 amino acids, preferably a length of 8 to 15 amino acids. 29. a) effector function competing antibodies that contain an Fc gamma receptor binding site and / or a C1q binding site located in either the CH2 and / or CH3 domain; b) an effector-negative antibody that comprises an Fc region that is deficient in binding to Fc gamma receptors and / or C1q; or c) The antibody according to any one of items 1 to 28, which comprises a pH-dependent FcRn-binding site located in either the CH2 and / or CH3 domain. 30. The antibody according to any of items 1 to 29, wherein the antibody specifically recognizes a first target by means of a monovalent binding site comprising a pair of heavy and light chains incorporating a binding site recognizing the target, wherein the light chain is bound to another light chain composed of a constant region, thereby forming an Fc region. 31. The antibody according to item 30, wherein the antibody is a bispecific antibody that specifically recognizes a second target, wherein the first target is TNF-α, CD3, CD16, CD47 or PD-1, and the second target is EGFR, ROR1 or HER2, HER3, Lag-3. 32. The antibody according to any of items 1 to 31, wherein the constant domains of the antibody are of human origin, humanized, or a functionally active variant thereof having at least 80% sequence identity to the respective human IgG1 antibody domains. 33. An isolated nucleic acid encoding the antibody according to any one of items 1 to 32. EXAMPLES
[0236] Example The following examples are offered by way of illustration and not by way of limitation.
[0237] 6.12.1. Example 1 Bivalent monospecific and bivalent bispecific constructs A bivalent monospecific B-Body recognizing TNFα was constructed using standard molecular biology techniques with the following configuration: VL(certolizumab)-CH3(knob)-CH2-CH3 / VH(certolizumab)-CH3(hole). In this construct: First Polypeptide Chain (SEQ ID NO:1) Domain A=VL (certolizumab) Domain B = CH3 (IgG1) (Knob: S354C + T366W) Domain D=CH2(IgG1) Domain E=CH3(IgG1) Second Polypeptide Chain (SEQ ID NO:2) Domain F=VH (certolizumab) Domain G = CH3 (IgG1) (hole: Y349C, T366S, L368A, Y407V) Third Polypeptide Chain: Identical to the first polypeptide chain Fourth Polypeptide Chain: Identical to the second polypeptide chain.
[0238] References to domains and polypeptide chains follow Figure 3. The overall organization of the construct is shown in Figure 4. The sequence of the first polypeptide chain having domain A, identified as abbreviated "(VL)", is provided in SEQ ID NO: 1. The sequence of the second polypeptide chain having domain F, identified as abbreviated "(VH)", is provided in SEQ ID NO: 2.
[0239] The full-length construct was expressed in an E. coli cell-free protein synthesis expression system for approximately 18 hours with gentle agitation at 26° C. After expression, the cell-free extract was centrifuged to pellet insoluble material, and the supernatant was diluted 2-fold with 10× kinetic buffer (Forte Bio) and used as a specimen for Biolayer interference.
[0240] Biotinylated TNFα was immobilized on a streptavidin sensor and subjected to a wave-shift response of about 1.5 nm. After establishing a baseline with 10× kinetic buffer, the sensor was immersed in the antibody construct analyte solution. The construct gave a response of about 3 nm, comparable to the conventional IgG format of certolizumab, demonstrating the ability of the bivalent monospecific construct to assemble into a functional full-length antibody. The results are shown in FIG. 5.
[0241] The inventors have also constructed a bivalent, bispecific antibody with the following domain organization: First polypeptide chain: VL-CH3-CH2-CH3 (knob) Second polypeptide chain: VH-CH3 Third polypeptide chain: VL-CL-CH2-CH3(hole) The fourth polypeptide chain is VH-CH1.
[0242] The sequences (excluding the variable region sequences) are shown in SEQ ID NO:3 (first polypeptide chain), SEQ ID NO:4 (second polypeptide chain), SEQ ID NO:5 (third polypeptide chain), and SEQ ID NO:6 (fourth polypeptide chain), respectively.
[0243] 6.12.2. Example 2 Bivalent, bispecific B-Body "BC1" We constructed a bivalent, bispecific construct, designated "BC1," specific for PD1 and a second antigen, "Antigen A." The salient features of the "BC1" construct are shown in FIG.
[0244] In more detail, with reference to the domains and polypeptide chains according to FIG. 3 and modifications from the native sequence indicated in brackets, the construction was as follows: First Polypeptide Chain (SEQ ID NO:8) Domain A = VL ("Antigen A") Domain B = CH3 (T366K; 445K, 446S, 447C tripeptide insert) Domain D=CH2 Domain E=CH3(T366W, S354C) Second polypeptide chain (SEQ ID NO:9): Domain F = VH ("Antigen A") Domain G = CH3 (L351D; 445G, 446E, 447C tripeptide insert) Third polypeptide chain (SEQ ID NO:10): Domain H=VL("Nivo") Domain I = CL (kappa) Domain J=CH2 Domain K=CH3(Y349C, D356E, L358M, T366S, L368A, Y407V) Fourth polypeptide chain (SEQ ID NO:11): Domain L=VH("Nivo") Domain M=CH1.
[0245] The A domain (SEQ ID NO:12) and the F domain (SEQ ID NO:16) form an antigen-binding site (A:F) specific for "antigen A." The H domain has a VH sequence derived from nivolumab, and the L domain has a VL sequence derived from nivolumab; H and L associate to form an antigen-binding site (H:L) specific for human PD1.
[0246] The B domain (SEQ ID NO: 13) has the sequence of human IgG1 CH3 with several mutations, namely the T366K, 445K, 446S and 447C insertions. The T366K mutation is the charge-pair cognate of the L351D residue in domain G. The "447C" residue in domain B is derived from the C-terminal KSC tripeptide insertion.
[0247] Domain D (SEQ ID NO: 14) has the sequence of human IgG1 CH2.
[0248] Domain E (SEQ ID NO: 15) has the sequence of human IgG1 CH3 with the mutations T366W and S354C. 366W is a "knob" mutation. 354C introduces a cysteine that can form a disulfide bond with the cognate 349C mutation in domain K.
[0249] Domain G (SEQ ID NO: 17) has the sequence of human IgG1 CH3 with the following mutations: L351D, and a 445G, 446E, 447C tripeptide insert. The L351D mutation introduces a charge pair cognate to the domain B T366K mutation. The "447C" residue in domain G is derived from the C-terminal GEC tripeptide insert.
[0250] Domain I (SEQ ID NO: 19) has the sequence of the human C kappa light chain (Cκ).
[0251] Domain J [SEQ ID NO: 20] has the sequence of the human IgG1 CH2 domain and is identical to the sequence of domain D.
[0252] Domain K [SEQ ID NO:21] has the sequence of human IgG1 CH3 with the following changes: Y349C, D356E, L358M, T366S, L368A, Y407V. The 349C mutation introduces a cysteine that can form a disulfide bond with the cognate 354C mutation in domain E. 356E and L358M introduce isoallotypic amino acids that reduce immunogenicity. 366S, 368A and 407V are "hole" mutations.
[0253] Domain M [SEQ ID NO: 23] has the sequence of the human IgG1 CH1 region.
[0254] "BC1" can be readily expressed at high levels, at concentrations in excess of 100 μg / ml, using mammalian expression.
[0255] The inventors have found that the bivalent, bispecific "BC1" protein can be easily purified in a single step using the CH1-specific CaptureSelect™ affinity resin from ThermoFisher.
[0256] As shown in Figure 7A, SEC analysis demonstrates that a single CH1 affinity purification step yields a single monodisperse peak that is >98% monomeric via gel filtration. Figure 7B shows comparative literature data of SEC analysis of CrossMab bivalent antibody constructs.
[0257] Figure 8A shows the cation exchange chromatography elution profile of "BC1" after one-step purification using CaptureSelect™ CH1 affinity resin, showing a single tight peak. Figure 8B shows the cation exchange chromatography elution profile of "BC1" after purification using standard Protein A purification, showing an additional elution peak consistent with the co-purification of an incompletely assembled product.
[0258] FIG. 9 shows an SDS-PAGE gel under non-reducing conditions. Single-step purification of "BC1" using CH1 affinity resin yields a nearly homogeneous single band as seen in lane 3. Lane 4 shows a subsequent minimal additional purification with a cation exchange polishing step. Lane 7 shows, by comparison, minor purification using standard Protein A purification, and lanes 8-10 show further purification of the Protein A purified material using cation exchange chromatography.
[0259] FIG. 10 shows an SDS-PAGE gel of "BC1" after single-step CH1-affinity purification under both non-reducing and reducing conditions (Panel A) compared with an SDS-PAGE gel of the CrossMab bispecific antibody under non-reducing and reducing conditions published in the literature (Panel B).
[0260] Figure 11 shows mass spectrometry analysis of "BC1" demonstrating two distinct heavy chains (Figure 11A) and two distinct light chains (Figure 11B) under reducing conditions. The mass spectrometry data in Figure 12 confirms the absence of mispairing after purification.
[0261] Accelerated stability studies were performed to evaluate the long-term stability of the "BC1" B-Body design. Purified B-Body was concentrated to 8.6 mg / ml in PBS buffer and incubated at 40° C. Structural integrity was measured weekly using analytical size-exclusion chromatography (SEC) with a Shodex KW-803 column. Structural integrity was determined by measuring the percentage of intact monomer (% monomer) in relation to aggregate formation. Data are shown in FIG. 13. IgG control 1 is a positive control with good stability properties. IgG control 2 is a negative control known to aggregate under the incubation conditions. The "BC1" B-Body was incubated for 8 weeks without any loss of structural integrity as determined by analytical SEC.
[0262] Furthermore, the TM of the bivalent construct was approximately 72° C., and the inventors determined that "BC1" has high thermostability.
[0263] Table 1 compares "BC1" with CrossMab in key developability characteristics. [Table 1]
[0264] 6.12.3. Example 3 Bivalent, bispecific B-Body "BC6" We constructed a bivalent, bispecific B-Body, designated "BC6." "BC6" is similar to "BC1," but is identical to BC1 except that it retains wild-type residues at residue 366 in domain B and residue 351 in domain G. Thus, "BC6" lacks the charge pair cognates T366K and L351D, which were designed to facilitate correct pairing in domains B and G of "BC1." The salient features of the "BC6" construct are shown in FIG. 14.
[0265] Despite the absence of charge-pair residues present in "BC1", single-step purification of "BC6" using CH1 affinity resin was found to result in a highly homogenous sample. Figure 15A shows the SEC analysis of "BC6" after one-step purification using CaptureSelect™ CH1 affinity resin. The data demonstrate that single-step CH1 affinity purification results in a single monodisperse peak, similar to that observed for "BC1", demonstrating that the disulfide bonds between polypeptide chains 1 and 2 and between polypeptide chains 3 and 4 remain intact. The chromatogram also shows the absence of non-covalent aggregates.
[0266] Figure 15B shows an SDS-PAGE gel under non-reducing conditions, lane 1 loaded with the first lot of "BC6" after single-step CH1 affinity purification, lane 2 loaded with the second lot of "BC6" after single-step CH1 affinity purification, lanes 3 and 4 demonstrate the further purification that can be achieved using ion exchange chromatography following CH1 affinity purification.
[0267] 6.12.4. Example 4 Bivalent, bispecific B-Body "BC28", "BC29", "BC30", "BC31" We constructed bivalent 1x1 bispecific B-Body "BC28", "BC29", "BC30" and "BC31" with engineered disulfides in the CH3 interface of domains B and G as alternative SS linkages to the C-terminal disulfides present in "BC1" and "BC6". The literature shows that the disulfide bond at the CH3 interface is insufficient to exert orthogonality in the context of the Fc CH3 domain. The overall organization of these B-Body constructs is outlined in Figure 16, and salient features of "BC28" are shown below: Polypeptide chain 1: "BC28" chain 1 (SEQ ID NO:24) Domain A = VL (antigen "A") Domain B = CH3 (Y349C; 445P, 446G, 447K inserts) Domain D=CH2 Domain E=CH3(S354C, K366W) Polypeptide chain 2: "BC28" chain 2 (SEQ ID NO:25) Domain F = VH (antigen "A") Domain G=CH3 (S354C; 445P, 446G, 447K inserts) Polypeptide chain 3: "BC1" chain 3 (SEQ ID NO: 10) Domain H=VL("Nivo") Domain I = CL (kappa) Domain J=CH2 Domain K=CH3(Y349C, D356E, L358M, T366S, L368A, Y407V) Polypeptide chain 4: "BC1" chain 4 (SEQ ID NO:11) Domain L=VH("Nivo") Domain M=CH1.
[0268] The "BC28" A:F antigen binding site is specific for "Antigen A". The "BC28" H:L antigen binding site is specific for PD1 (Nivolumab sequence). "BC28" domain B has the following changes compared to wild type CH3: Y349C; 445P, 446G, 447K insertions. "BC28" domain E has the following changes compared to wild type CH3: S354C, K366W. "BC28" domain G has the following changes compared to wild type: S354C; 445P, 446G, 447K insertions.
[0269] Thus, "BC28" has an engineered cysteine at residue 349C in domain B and an engineered cysteine at residue 354C in domain G ("349C-354C").
[0270] "BC29" has engineered cysteines at residues 351C in domain B and 351C in domain G ("351C-351C"). "BC30" has engineered cysteines at residues 354C in domain B and 349C in domain G ("354C-349C"). BC31 has an engineered cysteine at residue 394C and an engineered cysteine at 394C in domain G ("394C-394C"). BC32 has an engineered cysteine at residues 407C in domain B and 407C in domain G ("407C-407C").
[0271] FIG. 17 shows SDS-PAGE analysis under non-reducing conditions after one-step purification using CaptureSelect™ CH1 affinity resin. Lanes 1 and 3 show high levels of expression and substantial homogeneity of intact “BC28” (lane 1) and “BC30” (lane 3). Lane 2 shows oligomerization of BC29. Lanes 4 and 5 show poor expression of BC31 and BC32, respectively, indicating poor ligation of BC32. Another construct, BC9 (a construct reported by Genentech with cysteines introduced at residues 392 in domain B and 399 in domain G (“392C-399C”) to form a disulfide pair) showed oligomerization on SDS PAGE (data not shown).
[0272] Figure 18 shows the SEC analysis of "BC28" and "BC30" after one-step purification using CaptureSelect™ CH1 affinity resin. Furthermore, we demonstrated that "BC28" can be easily purified using a single-step purification using Protein A resin (results not shown).
[0273] 6.12.5. Example 5 Bivalent, bispecific B-Body "BC44" FIG. 19 shows the general organization of our currently preferred bivalent, bispecific 1×1 construct, the bivalent, bispecific 1×1 B-Body “BC44”. First Polypeptide Chain ("BC44" Chain 1) (SEQ ID NO:32) Domain A = VL (antigen "A") Domain B = CH3 (P343V; Y349C; 445P, 446G, 447K insertions) Domain E=CH2 Domain E=CH3(S354C, K366W) Second Polypeptide Chain (= "BC28" Polypeptide Chain 2) (SEQ ID NO: 25) Domain F = VH (antigen "A") Domain G=CH3 (S354C; 445P, 446G, 447K inserts) Third Polypeptide Chain (="BC1" Polypeptide Chain 3) (SEQ ID NO: 10) Domain H=VL("Nivo") Domain I = CL (kappa) Domain J=CH2 Domain K=CH3(Y349C, D356E, L358M, T366S, L368A, Y407V) Fourth Polypeptide Chain (="BC1" Polypeptide Chain 4) (SEQ ID NO:11) Domain L=VH("Nivo") Domain M=CH1
[0274] 6.12.6. Example 6 Variable-CH3 Junction Operation We generated a series of variants in which the VL-CH3 junction between domains A and B, and the VH-CH3 junction between domains F and G were mutated, and evaluated the expression levels, assembly, and stability of the bivalent 1x1 B-Body constructs. While many solutions are possible, we chose to use only residues naturally found in the VL, VH, and CH3 domains to reduce the introduction of T cell epitopes. Structural evaluation of the domain configurations further qualifies the preferred sequence combinations. Tables 2 and 3 below show the junctions for several junction variants based on "BC1" and other bivalent constructs. [Table 2] [Table 3]
[0275] Figure 20 shows size exclusion chromatography of "BC15" and "BC16" samples at the indicated weeks in the accelerated stability testing protocol at 40° C. "BC15" remained stable; "BC16" was found to become unstable over time.
[0276] 6.12.7. Example 7 Trivalent 2x1 Bispecific B-Body Construct ("BC1-2x1") Based on "BC1", we constructed a trivalent 2x1 bispecific B-Body "BC1-2x1". The salient features of the construct are shown in Figure 22.
[0277] More specifically, with reference to the domains and polypeptide chains summarized in FIG. First Polypeptide Chain Domain N=VL ("Antigen A") Domain O=CH3(T366K, 447C) Domain A = VL ("Antigen A") Domain B = CH3 (T366K, 477C) Domain D=CH2 Domain E=CH3 (Knob, 354C) The fifth polypeptide chain (= "BC1" chain 2) Domain P=VH ("Antigen A") Domain Q=CH3(L351D, 447C) Second Polypeptide Chain (= "BC1" Chain 2) Domain F = VH ("Antigen A") Domain G=CH3(L351D, 447C) Third Polypeptide Chain (= "BC1" Chain 3) Domain H=VL("Nivo") Domain I = CL (kappa) Domain J=CH2 Domain K=CH3 (Hole, 349C) The fourth polypeptide chain (= "BC1" chain 4) Domain L=VH("Nivo") Domain M=CH1
[0278] FIG. 23 shows non-reducing SDS-PAGE of proteins expressed using the ThermoFisher Expi293 transient transfection system.
[0279] Lane 1 shows the eluate of the trivalent 2x1 "BC1-2x1" protein after one-step purification using CaptureSelect™ CH1 affinity resin. Lane 2 shows the lower molecular weight, faster migrating, bivalent "BC1" protein after one-step purification using CaptureSelect™ CH1 affinity resin. Lanes 3-5 show purification of "BC1-2x1" using Protein A. Lanes 6 and 7 show purification of "BC1-2x1" using CH1 affinity resin.
[0280] Figure 24 compares the avidity of the bivalent "BC1" construct to that of the trivalent 2x1 "BC1-2x1" construct using Octet (Pall ForteBio) analysis. Biotinylated antigen "A" is immobilized on a surface over which the antibody construct is passed for binding analysis.
[0281] 6.12.8. Example 8 Trivalent 2x1 Trispecific B-Body Construct ("TB111") The inventors have designed a trivalent 2x1 trispecific molecule "TB111" with the structure outlined in Figure 25. With reference to the domain naming convention shown in Figure 21, TB111 has the following structure ("Ada" indicates the V region derived from adalimumab): Polypeptide chain 1 Domain N:VH("Ada") Domain O: CH3 (T366K, 394C) Domain A: VL ("Antigen A") Domain B: CH3 (T366K, 349C) Domain D:CH2 Domain E: CH3 (knob, 354C) Polypeptide chain 5 Domain P:VL("Ada") Domain Q: CH3 (L351D, 394C) Polypeptide chain 2 Domain F: VH ("Antigen A") Domain G: CH3 (L351D, 351C) Polypeptide chain 3 Domain H:VL ("Nivo") Domain I: CL (kappa) Domain J:CH2 Domain K: CH3 (Hole, 349C) Polypeptide chain 4 (= "BC1" chain 4) Domain L: VH ("Nivo") Domain M:CH1 This construct was not expressed.
[0282] 6.12.9. Example 9 Trivalent 1x2 bispecific construct ("BC28-1x2") We constructed a trivalent 1x2 bispecific B-Body with the following domain structure: First Polypeptide Chain (= "BC28" Chain 1) (SEQ ID NO: 24) Domain A = VL (antigen "A") Domain B = CH3 (Y349C; 445P, 446G, 447K inserts) Domain D=CH2 Domain E=CH3(S354C, K366W) Second polypeptide chain (= "BC28" chain 2) (SEQ ID NO: 25) Domain F = VH (antigen "A") Domain G=CH3 (S354C; 445P, 446G, 447K inserts) Third Polypeptide Chain (SEQ ID NO:37) Domain R=VL (antigen "A") Domain S=CH3 (Y349C; 445P, 446G, 447K inserts) Linker=GSGSGS Domain H=VL("Nivo") Domain I=CL Domain J=CH2 Domain K=CH3(Y349C, D356E, L358M, T366S, L368A, Y407V) Fourth polypeptide chain (= "BC1" chain 4) (SEQ ID NO:11): Domain L=VH("Nivo") Domain M=CH1. The sixth polypeptide chain (= "BC28" chain 2) (SEQ ID NO: 25) Domain T = VH (antigen "A") Domain U=CH3 (S354C; 445P, 446G, 447K inserts)
[0283] Similar to the H:L linked antigen binding site, the A:F antigen binding site is specific for "antigen A". The R:T antigen binding site is specific for PD. The specificity of this construct is therefore antigen "A" x (PD1-antigen "A").
[0284] 6.12.10. Example 10 Trivalent 1x2 bispecific construct ("CTLA4-4xNivoxCTLA4-4") We constructed a trivalent 1x2 bispecific molecule ("CTLA4-4xNivoxCTLA4-4") with the general structure outlined in Figure 27. The domain nomenclature is shown in Figure 26.
[0285] FIG. 28 is an SDS-PAGE gel in which the lanes showing the "CTLA4-4 x Nivo x CTLA4-4" construct under non-reducing and reducing conditions are boxed.
[0286] Figure 29 compares the antigen binding of the two antibodies "CTLA4-4xOX40-8" and "CTLA4-4xNivoxCTLA4-4." "CTLA4-4xOX40-8" binds monovalently to CTLA4, while "CTLA4-4xNivoxCTLA4-4" binds bivalently to CTLA4.
[0287] 6.12.11. Example 11 Trivalent 1x2 trispecific construct "BC28-1x1x1a" The inventors have constructed a trivalent 1x2 trispecific molecule with the general structure outlined in Figure 30. With reference to the domain nomenclature shown in Figure 26, First polypeptide chain (= "BC28" chain 1) [SEQ ID NO: 24] Domain A = VL (antigen "A") Domain B = CH3 (Y349C; 445P, 446G, 447K inserts) Domain D=CH2 Domain E=CH3(S354C, K366W) Second polypeptide chain (= "BC28" chain 2) (SEQ ID NO: 25) Domain F = VH (antigen "A") Domain G=CH3 (S354C; 445P, 446G, 447K inserts) Third Polypeptide Chain (SEQ ID NO:45) Domain R=VL(CTLA4-4) Domain S=CH3 (T366K; 445K, 446S, 447C insert) Linker=GSGSGS Domain H=VL("Nivo") Domain I=CL Domain J=CH2 Domain K=CH3(Y349C, D356E, L358M, T366S, L368A, Y407V) Fourth polypeptide chain (= "BC1" chain 4) (SEQ ID NO: 11) Domain L=VH("Nivo") Domain M=CH1. Sixth Polypeptide Chain (=hCTLA4-4 Chain 2) (SEQ ID NO:53) Domain T=VH(CTLA4) Domain U=CH3 (L351D, 445G, 446E, 447C inserts)
[0288] The antigen binding sites of this trispecific construct were as follows: The antigen-binding site A:F was specific for “antigen A”; The antigen-binding site H:L was specific for PD1 (nivolumab sequence); The antigen-binding site R:T was specific for CTLA4.
[0289] FIG. 31 shows size-exclusion chromatography of "BC28-1x1x1a" after transient expression and one-step purification using CaptureSelect™ CH1 affinity resin, showing a single clearly defined peak.
[0290] 6.12.12. Example 12 SDS-PAGE analysis of bivalent and trivalent constructs FIG. 32 shows SDS-PAGE gels under non-reducing and reducing conditions of various constructs after transient expression and one-step purification using CaptureSelect™ CH1 affinity resin, respectively.
[0291] Lanes 1 (non-reduced conditions) and 2 (reduced conditions, +DTT) are the bivalent 1x1 bispecific construct "BC1". Lanes 3 (non-reduced) and 4 (reduced) are the trivalent bispecific 2x1 construct "BC1-2x1" (see Example 7). Lanes 5 (non-reduced) and 6 (reduced) are the trivalent 1x2 bispecific construct "CTLA4-4xNivoxCTLA4-4" (see Example 10). Lanes 7 (non-reduced) and 8 (reduced) are the trivalent 1x2 trispecific "BC28-1x1x1a" construct described in Example 11.
[0292] SDS-PAGE gels demonstrate complete assembly of each construct, and for each construct the major band on the non-reducing gel is shown at the expected molecular weight.
[0293] 6.12.13. Example 13 Binding analysis Figure 33 shows Octet binding analysis to three antigens: PD1, antigen "A" and CTLA-4. In each case, the antigen is immobilized and the B-Body is the specimen. For reference, the 1x1 bispecifics "BC1" and "CTLA4-4xOX40-8" are also compared, demonstrating that the 1x1 B-Body specifically binds only to antigens for which the antigen binding site is selected.
[0294] Figure 33A shows that "BC1" binds to PD1 and antigen "A" but not to CTLA4. Figure 33B shows that the bivalent bispecific 1x1 construct "CTLA4-4xOX40-8" binds to CTLA4 but not to antigen "A" or PD1. Figure 33C shows that the trivalent trispecific 1x2 construct "BC28-1x1x1a" binds to PD1, antigen "A" and CTLA4.
[0295] 6.12.14. Example 14 Tetravalent constructs Figure 35 shows the overall organization of the 2x2 tetravalent bispecific construct "BC22-2x2". The 2x2 tetravalent bispecific was constructed on the "BC1" scaffold by duplicating each variable domain-constant domain segment. The domain nomenclature is outlined in Figure 34.
[0296] Figure 36 is an SDS-PAGE gel. Lanes 7-9 show the "BC22-2x2" tetravalent construct after one-step purification using CaptureSelect™ CH1 affinity resin ("CH1 eluate") and after further ion exchange chromatography purification (lane 8, "pk1 after IEX"; lane 9, "pk2 after IEX"). Lanes 1-3 are the trivalent 2x1 construct "BC21-2x1" after CH1 affinity purification (lane 1), lanes 2 and 3 after further ion exchange chromatography. Lanes 4-6 are the 1x2 trivalent construct "BC12-1x2".
[0297] FIG. 37 shows the overall organization of the 2×2 tetravalent construct.
[0298] Figures 39 and 40 are schematic diagrams of tetravalent constructs with alternative configurations. The domain nomenclature is shown in Figure 38.
[0299] 6.12.15. Example 15 Bispecific antigen engagement by B-Body A tetravalent bispecific 2x2 B-Body "B-Body-IgG 2x2" was constructed. More specifically, using the domain and polypeptide chain references summarized in Figure 38, First Polypeptide Chain Domain A=VL (certolizumab) Domain B=CH3 (IgG1, Knob) Domain D=CH2(IgG1) Domain E=CH3(IgG1) Domain W = VH (antigen "A") Domain X=CH1(IgG1) a third polypeptide chain (identical to the first polypeptide chain) Domain H=VL (certolizumab) Domain I = CH3 (IgG1, Knob) Domain J=CH2(IgG1) Domain K=CH3(IgG1) Domain WW = VH (antigen "A") Domain XX=CH1(IgG1) Second Polypeptide Chain Domain F=VH (certolizumab) Domain G=CH3 (IgG1, whole) A fourth polypeptide chain (identical to the third polypeptide chain) Domain F=VH (certolizumab) Domain G=CH3 (IgG1, whole) Seventh Polypeptide Chain Domain Y = VH ("Antigen A") Domain Z=CL Kappa Eighth polypeptide chain (identical to the seventh polypeptide chain) Domain YY=VH ("Antigen A") Domain ZZ=CL Kappa.
[0300] It was cloned and expressed as described in Example 1. Here, the BLI experiment consisted of immobilization of biotinylated antigen "A" onto a streptavidin sensor followed by establishment of a baseline using 10x kinetic buffer. The sensor was then immersed in cell-free expressed "B-Body-IgG 2x2" and then a new baseline was established. Finally, the sensor was immersed in 100 nM TNFα where a second binding event was observed, confirming bispecific binding of both antigens by a single "B-Body-IgG 2x2" construct. The results are shown in Figure 41.
[0301] 6.12.16. (Example 16) Antigen-specific cell binding of "BB-IgG 2×2" Expi-293 cells were mock transfected or transiently transfected with antigen "B" using the Expi-293 transfection kit (Life Technologies). 48 hours after transfection, Expi-293 cells were harvested and fixed in 4% paraformaldehyde for 15 minutes at room temperature. Cells were washed twice in PBS. 200,000 antigen B or mock transfected Expi-293 cells were placed in 100uL of PBS in a V-bottom 96-well plate. Cells were incubated with "B-Body-IgG 2x2" at a concentration of 3ug / mL for 1.5 hours at room temperature. Cells were centrifuged at 300xG for 7 minutes, washed with PBS, and incubated with 100uL of FITC-labeled goat anti-human secondary antibody at a concentration of 8ug / mL for 1 hour at room temperature. The cells were centrifuged at 300×G for 7 min, washed with PBS, and cell binding was confirmed by flow cytometry using a Guava easyCyte. The results are shown in FIG.
[0302] 6.12.17. (Example 17) SDS-PAGE analysis of bivalent and trivalent constructs FIG. 45 shows SDS-PAGE gels under non-reducing and reducing conditions of various constructs after transient expression and one-step purification using CaptureSelect™ CH1 affinity resin, respectively.
[0303] Lanes 1 (non-reduced conditions) and 2 (reduced conditions, +DTT) are the bivalent 1x1 bispecific construct "BC1". Lanes 3 (non-reduced) and 4 (reduced) are the bivalent 1x1 bispecific construct "BC28" (see Example 4). Lanes 5 (non-reduced) and 6 (reduced) are the bivalent 1x1 bispecific construct "BC44" (see Example 5). Lanes 7 (non-reduced) and 8 (reduced) are the trivalent 1x2 bispecific "BC28-1x2" construct (see Example 9). Lanes 9 (non-reduced) and 10 (reduced) are the trivalent 1x2 trispecific "BC28-1x1x1a" construct described in Example 11.
[0304] SDS-PAGE gels demonstrate complete assembly of each construct, and for each construct the major band on the non-reducing gel is shown at the expected molecular weight.
[0305] 6.12.18. (Example 18) Stability analysis of variable-CH3 junction operation The pairing stability between various junction variant combinations was evaluated. Differential scanning fluorimetry was performed to determine the melting temperatures of various junction variant pairings between the VL-CH3 polypeptide of chain 1 (domains A and B) and the VH-CH3 polypeptide of chain 2 (domains F and G). The junction variants "BC6jv", "BC28jv", "BC30jv", "BC44jv" and "BC45jv" (with the corresponding junction sequences of "BC6", "BC28", "BC30", "BC44" and "BC45" seen in Tables 2 and 3 above, respectively) have Tm in the range of 76-77 degrees, demonstrating increased pairing stability (see Table 4). Figure 46 shows the difference in thermal transitions of "BC24jv", "BC26jv" and "BC28jv", demonstrating that "BC28jv" has the greatest stability among the three. The x-axis of the figure is temperature and the y-axis is the change in fluorescence divided by the change in temperature (-dFluor / dTemp). Experiments were performed as described in Niesen et al. (Nature Protocols, (2007) 2:2212-2221), which is incorporated herein by reference for all its teachings. [Table 4] Arrays [ka] [ka] [ka] [ka] [ka] [ka] [ka] [ka] [ka] [ka] [ka] [ka] [ka] [ka] Other sequences: > Hinge: DKTHTCPPCP [SEQ ID NO: 56] >BC1-Polypeptide 1 Domain Junction: IKRTPREP [SEQ ID NO: 57] >BC15-polypeptide 1 domain junction: IKRTVREP [SEQ ID NO: 58] >BC16-polypeptide 1 domain junction: IKRTREP [SEQ ID NO: 59] >BC17-polypeptide 1 domain junction: IKRTVPREP [SEQ ID NO: 60] >BC26-polypeptide 1 domain junction: IKRTVAEP [SEQ ID NO: 61] >BC27-polypeptide 1 domain junction: IKRTVAPREP [SEQ ID NO: 62] >BC1-polypeptide 2 domain junction: SSASPREP [SEQ ID NO: 63] >BC13-polypeptide 2 domain junction: SSASTREP [SEQ ID NO: 64] >BC14-polypeptide 2 domain junction: SSASTPREP [SEQ ID NO: 65] >BC24-polypeptide 2 domain junction: SSASTKGEP [SEQ ID NO: 66] >BC25-polypeptide 2 domain junction: SSASTKGREP [SEQ ID NO: 67]
[0306] 7. INCORPORATION BY REFERENCE All publications, patents, patent applications, and other documents cited in this application are incorporated by reference in their entirety herein for all purposes to the same extent as if each individual publication, patent, patent application, or other document was individually indicated to be incorporated by reference for all purposes.
[0307] 8. Equivalents While various specific embodiments have been illustrated and described, the above specification is not intended to be limiting. It is understood that various modifications may be made without departing from the spirit and scope of the invention. Many variations will become apparent to those skilled in the art upon review of this specification. The present invention provides, for example, the following items. (Item 1) A binding molecule comprising a first and a second polypeptide chain, (a) the first polypeptide chain comprises Domain A, Domain B, Domain D, and Domain E; wherein the domains are arranged N-terminus to C-terminus in an ABDE orientation; Domain A has a VL amino acid sequence, Domain B has a CH3 amino acid sequence, Domain D has a CH2 amino acid sequence, and Domain E has a constant region domain amino acid sequence; (b) the second polypeptide chain comprises Domain F and Domain G; wherein the domains are arranged N-terminus to C-terminus in an FG orientation; Domain F has a VH amino acid sequence and Domain G has a CH3 amino acid sequence; (c) the first and second polypeptides associate via an interaction between the A domain and the F domain and an interaction between the B domain and the G domain to form the binding molecule. (Item 2) further comprising a third and a fourth polypeptide chain; (a) the third polypeptide chain comprises domain H, domain I, domain J, and domain K; wherein the domains are arranged N-terminus to C-terminus in the orientation HIJK; Domain H has a variable region domain amino acid sequence, Domain I has a constant region domain amino acid sequence, Domain J has a CH2 amino acid sequence, and Domain K has a constant region domain amino acid sequence; (b) the fourth polypeptide chain comprises Domain L and Domain M; wherein the domains are arranged N-terminus to C-terminus in LM orientation; Domain L has a variable region domain amino acid sequence and Domain M has a constant region amino acid sequence; (c) the third and fourth polypeptides associate through an interaction between the H domain and the L domain and an interaction between the I domain and the M domain; (d) the first and third polypeptides associate via an interaction between the D domain and the J domain and an interaction between the E domain and the K domain to form the binding molecule. (Item 3) 3. The binding molecule of any one of items 1 or 2, wherein the amino acid sequences of the B domain and the G domain are identical, and the sequences are endogenous CH3 sequences. (Item 4) 3. The binding molecule of any one of items 1 or 2, wherein the amino acid sequences of the B domain and the G domain are distinct and each separately comprises an orthogonal modification to an endogenous CH3 sequence, wherein the B domain interacts with the G domain, and neither the B domain nor the G domain interacts significantly with a CH3 domain lacking the orthogonal modification. (Item 5) 5. The binding molecule according to item 4, wherein the orthogonal modification comprises a mutation that creates an engineered disulfide bridge between domains B and G. (Item 6) 6. The binding molecule according to item 5, wherein the mutations generating engineered disulfide bridges are an S354C mutation in one of the B and G domains and a 349C mutation in the other domain. (Item 7) 7. The binding molecule according to any one of items 4 to 6, wherein the orthogonal modification comprises a knob-in-hole mutation. (Item 8) 8. The binding molecule according to item 7, wherein the knob-in-hole mutation is a T366W mutation in one of the B and G domains and a T366S, L368A and Y407V mutation in the other domain. (Item 9) 9. The binding molecule of any one of items 4 to 8, wherein the orthogonal modification comprises a charge pair mutation. (Item 10) 10. The binding molecule of item 9, wherein the charge pair mutation is a T366K mutation in one of the B and G domains and a L351D mutation in the other domain. (Item 11) 11. The binding molecule according to any one of items 1 to 10, wherein domain E has a CH3 amino acid sequence. (Item 12) 12. The binding molecule according to any one of items 2 to 11, wherein the amino acid sequences of the E and K domains are identical, said sequences being endogenous CH3 sequences. (Item 13) 12. The binding molecule according to any one of items 2 to 11, wherein the amino acid sequences of the E and K domains are different. (Item 14) 14. The binding molecule of item 13, wherein the different sequences each independently comprise an orthogonal modification to the endogenous CH3 sequence, wherein the E domain interacts with the K domain, and wherein neither the E domain nor the K domain interacts significantly with a CH3 domain lacking the orthogonal modification. (Item 15) 15. The binding molecule of item 14, wherein the orthogonal modification comprises a mutation that creates an engineered disulfide bridge between domains E and K. (Item 16) 16. The binding molecule according to item 15, wherein the mutations generating engineered disulfide bridges are an S354C mutation in one of the E and K domains and a 349C mutation in the other domain. (Item 17) 17. The binding molecule according to any one of items 14 to 16, wherein the orthogonal modifications in the E and K domains comprise knobs-in-hole mutations. (Item 18) 18. The binding molecule according to item 17, wherein the knob-in-hole mutation is a T366W mutation in one of the E or K domains and a T366S, L368A and Y407V mutation in the other domain. (Item 19) 19. The binding molecule of any one of items 14 to 18, wherein the orthogonal modification comprises a charge pair mutation. (Item 20) 20. The binding molecule of item 19, wherein the charge pair mutation is a T366K mutation in one of the E or K domains and a corresponding L351D mutation in the other domain. (Item 21) The binding molecule of item 13, wherein the amino acid sequences of the E domain and the K domain are endogenous sequences of two different antibody domains selected to have specific interactions that promote a specific association between the first polypeptide and the third polypeptide. (Item 22) 22. The binding molecule according to item 21, wherein the two different amino acid sequences are a CH1 sequence and a CL sequence. (Item 23) 23. The binding molecule according to any one of items 2 to 22, wherein domain I has a CL sequence and domain M has a CH1 sequence. (Item 24) 24. The binding molecule according to any one of items 2 to 23, wherein domain H has a VL sequence and domain L has a VH sequence. (Item 25) Domain H has the VL amino acid sequence; Domain I has the amino acid sequence CL; Domain K has a CH3 amino acid sequence; Domain L has a VH amino acid sequence; 25. The binding molecule according to any one of items 2 to 24, wherein domain M has a CH1 amino acid sequence. (Item 26) 26. The binding molecule of any one of items 2 to 25, wherein the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen, and the interaction between the H domain and the L domain forms a second antigen-binding site specific for a second antigen. (Item 27) 1. A binding molecule comprising a first, second, third, and fourth polypeptide chain: (a) the first polypeptide chain comprises Domain A, Domain B, Domain D, and Domain E; wherein the domains are arranged N-terminus to C-terminus in an ABDE orientation; Domain A has a VL amino acid sequence, Domain B has a CH3 amino acid sequence, Domain D has a CH2 amino acid sequence, and Domain E has a constant region domain amino acid sequence; (b) the second polypeptide chain comprises Domain F and Domain G; wherein the domains are arranged N-terminus to C-terminus in an FG orientation; Domain F has a VH amino acid sequence and Domain G has a CH3 amino acid sequence; (c) the third polypeptide chain comprises domain H, domain I, domain J, and domain K; wherein the domains are arranged N-terminus to C-terminus in the orientation HIJK; Domain H has a variable region domain amino acid sequence, Domain I has a constant region domain amino acid sequence, Domain J has a CH2 amino acid sequence, and Domain K has a constant region domain amino acid sequence; (d) the fourth polypeptide chain comprises Domain L and Domain M; wherein the domains are arranged N-terminus to C-terminus in LM orientation; Domain L has a variable region domain amino acid sequence and Domain M has a constant region domain amino acid sequence; (e) the first and second polypeptides associate via an interaction between the A domain and the F domain and an interaction between the B domain and the G domain; (f) the third and fourth polypeptides associate through an interaction between the H domain and the L domain and an interaction between the I domain and the M domain; (g) the first and third polypeptides associate via an interaction between the D domain and the J domain and an interaction between the E domain and the K domain to form a binding molecule. (Item 28) the E domain has a CH3 amino acid sequence; Domain H has the VL amino acid sequence; Domain I has the amino acid sequence CL; Domain K has a CH3 amino acid sequence; Domain L has a VH amino acid sequence; 28. The binding molecule according to item 27, wherein domain M has a CH1 amino acid sequence. (Item 29) further comprising a fifth polypeptide chain, (a) the first polypeptide chain further comprises Domain N and Domain O, wherein the domains are arranged, from N-terminus to C-terminus, in a NOABDE orientation; Domain N has a VL amino acid sequence and Domain O has a CH3 amino acid sequence; (b) the binding molecule further comprises a fifth polypeptide chain, the fifth polypeptide chain comprising Domain P and Domain Q; wherein the domains are arranged N-terminus to C-terminus in a PQ orientation; Domain P has a VH amino acid sequence and domain Q has a CH3 amino acid sequence; (c) the first and fifth polypeptides associate via an interaction between the N domain and the P domain and an interaction between the O domain and the Q domain to form the binding molecule. (Item 30) (a) the amino acid sequences of domain N and domain A are identical; the amino acid sequence of domain H is different from the amino acid sequences of domains N and A; The amino acid sequences of domain O and domain B are identical; the amino acid sequence of domain I is different from the amino acid sequences of domains O and B; The amino acid sequences of domain P and domain F are identical; the amino acid sequence of domain L is different from the amino acid sequences of domains P and F; The amino acid sequences of domain Q and domain G are identical; the amino acid sequence of domain M differs from the amino acid sequences of domains Q and G; (b) the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen, the interaction between the H domain and the L domain forms a second antigen-binding site specific for a second antigen, and the N and P domains form a third antigen-binding site specific for the first antigen. (Item 31) (a) the amino acid sequences of domains N, A, and H are different; The amino acid sequences of domains O, B, and I are different. The amino acid sequences of domains P, F, and L are different; The amino acid sequences of domains Q, G, and M are different; (b) the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen; the interaction between the H domain and the L domain forms a second antigen-binding site specific for a second antigen; 30. The binding molecule of item 29, wherein domain N and domain P form a third antigen-binding site specific for a third antigen. (Item 32) further comprising a sixth polypeptide chain, (a) the third polypeptide chain further comprises a domain R and a domain S; wherein the domains are arranged, from N-terminus to C-terminus, in the orientation RSHIJK; Domain R has a VL amino acid sequence and domain S has a constant domain amino acid sequence; (b) the binding molecule further comprises a sixth polypeptide chain, the sixth polypeptide chain comprising a domain T and a domain U; wherein the domains are arranged N-terminus to C-terminus in TU orientation; Domain T has a VH amino acid sequence and domain U has a constant domain amino acid sequence; 27. The binding molecule of any one of items 2 to 26, wherein (c) the third and sixth polypeptides associate via an interaction between the R domain and the T domain and an interaction between the S domain and the U domain to form the binding molecule. (Item 33) (a) the amino acid sequences of domain R and domain A are identical; The amino acid sequence of domain H is different from the amino acid sequences of domains R and A, The amino acid sequences of domain S and domain B are identical; the amino acid sequence of domain I is different from the amino acid sequences of domains S and B; The amino acid sequences of domain T and domain F are identical; the amino acid sequence of domain L is different from the amino acid sequences of domains T and F; The amino acid sequences of domain U and domain G are identical; the amino acid sequence of domain M differs from the amino acid sequences of domains U and G; (b) the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen; the interaction between the H domain and the L domain forms a second antigen-binding site specific for a second antigen; 33. The binding molecule of item 32, wherein the domain R and the domain T form a third antigen-binding site specific for the first antigen. (Item 34) (a) the amino acid sequences of domain R and domain H are identical; The amino acid sequence of domain A is different from the amino acid sequences of domains R and H, The amino acid sequences of domain S and domain I are identical; the amino acid sequence of domain B is different from the amino acid sequences of domains S and I; The amino acid sequences of domain T and domain L are identical; the amino acid sequence of domain F differs from the amino acid sequences of domains T and L; The amino acid sequences of domain U and domain M are identical; the amino acid sequence of domain G differs from the amino acid sequences of domains U and M; (b) the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen; the interaction between the H domain and the L domain forms a second antigen-binding site specific for a second antigen; 33. The binding molecule of claim 32, wherein the R and T domains form a third antigen-binding site specific for the second antigen. (Item 35) (a) the amino acid sequences of domains R, A, and H are different; The amino acid sequences of domains S, B, and I are different. The amino acid sequences of domains T, F, and L are different; The amino acid sequences of domains U, G, and M are different; (b) the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen; The binding molecule of claim 32, wherein the interaction between the H domain and the L domain forms a second antigen-binding site specific for a second antigen, and the R and T domains form a third antigen-binding site specific for a third antigen. (Item 36) further comprising a fifth and a sixth polypeptide chain, (a) the first polypeptide chain further comprises Domain N and Domain O, wherein the domains are arranged from N-terminus to C-terminus in a NOABDE orientation; (b) the third polypeptide chain further comprises a domain R and a domain S, wherein the domains are arranged, from N-terminus to C-terminus, in the following orientation: RSHIJK; (c) the binding molecule further comprises fifth and sixth polypeptide chains: the fifth polypeptide chain comprises a domain P and a domain Q, wherein the domains are arranged, from N-terminus to C-terminus, in a PQ orientation; the sixth polypeptide chain comprises a domain T and a domain U, wherein the domains are arranged N-terminus to C-terminus in a TU orientation; (d) the first and fifth polypeptides associate through an interaction between the N domain and the P domain and an interaction between the O domain and the Q domain; 27. The binding molecule of any one of items 2 to 26, wherein the third and sixth polypeptides associate via an interaction between the R domain and the T domain and an interaction between the S domain and the U domain to form the binding molecule. (Item 37) (a) the amino acid sequences of domain N and domain A are identical; The amino acid sequences of domain H and domain R are identical; The amino acid sequences of domain O and domain B are identical; The amino acid sequences of domain I and domain S are identical; The amino acid sequences of domain P and domain F are identical; The amino acid sequences of domain L and domain T are identical; The amino acid sequences of domain Q and domain G are identical; the amino acid sequences of domain M and domain U are identical; (b) The binding molecule of item 36, wherein the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen, the N domain and the P domain form a second antigen-binding site specific for the first antigen, the interaction between the H domain and the L domain forms a third antigen-binding site specific for a second antigen, and the interaction between the R domain and the T domain forms a fourth antigen-binding site specific for the second antigen. (Item 38) (a) the amino acid sequences of domain H and domain A are identical; The amino acid sequences of domain N and domain R are identical; The amino acid sequences of domain I and domain B are identical; The amino acid sequences of domain O and domain S are identical; The amino acid sequences of domain L and domain F are identical; The amino acid sequences of domain P and domain T are identical; The amino acid sequences of domain M and domain G are identical; the amino acid sequences of domain Q and domain U are identical; (b) The binding molecule of item 36, wherein the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen, the N domain and the P domain form a second antigen-binding site specific for a second antigen, the interaction between the H domain and the L domain forms a third antigen-binding site specific for the first antigen, and the interaction between the R domain and the T domain forms a fourth antigen-binding site specific for the second antigen. (Item 39) The binding molecule according to any of the preceding items, wherein the sequence forming the junction between the A domain and the B domain is IKRTPREP or IKRTVREP. (Item 40) The binding molecule according to any of the preceding items, wherein the sequence forming the junction between the F domain and the G domain is SSASPREP. (Item 41) The binding molecule of any of the preceding items, wherein at least one CH3 amino acid sequence has a C-terminal tripeptide insertion linking the CH3 amino acid sequence and a hinge amino acid sequence, wherein the tripeptide insertion is selected from the group consisting of PGK, KSC and GEC. (Item 42) The binding molecule of any of the preceding items, wherein the sequence is a human sequence. (Item 43) The binding molecule of any of the preceding items, wherein at least one CH3 amino acid sequence is an IgG sequence. (Item 44) 44. The binding molecule of item 43, wherein the IgG sequence is an IgG1 sequence. (Item 45) The binding molecule of any of the preceding items, wherein at least one CH3 amino acid sequence has one or more isoallotypic mutations. (Item 46) 46. The binding molecule of item 45, wherein the isoallotypic mutation is D356E and L358M. (Item 47) The CL amino acid sequence is C カッパItem 2. The binding molecule of any of the preceding items, which is an array. (Item 48) A binding molecule according to any of the preceding items, A pharma- ceutically acceptable carrier; 13. A pharmaceutical composition comprising: (Item 49) A method of treatment comprising administering the pharmaceutical composition according to item 48 to a subject in need of treatment.
Claims
1. 1. An antibody construct comprising a first and a second polypeptide chain, (a) the first polypeptide chain comprises Domain A, Domain B, Domain D, and Domain E; wherein the domains are arranged, from N-terminus to C-terminus, in the orientation A-B-D-E; Domain A has a VL amino acid sequence, Domain B has a CH3 amino acid sequence, Domain D has a CH2 amino acid sequence, and Domain E has a constant region domain amino acid sequence; (b) the second polypeptide chain comprises Domain F and Domain G; wherein the domains are arranged, N-terminus to C-terminus, in an F-G orientation; Domain F has a VH amino acid sequence and Domain G has a CH3 amino acid sequence; (c) the first and second polypeptide chains associate via interactions between the A domain and the F domain and between the B domain and the G domain to form the antibody construct; the antibody construct further comprises a third and a fourth polypeptide chain; (d) the third polypeptide chain comprises domain H, domain I, domain J, and domain K; wherein the domains are arranged, from N-terminus to C-terminus, in a H-I-J-K orientation; Domain H has a VL amino acid sequence, Domain I has a CL amino acid sequence, Domain J has a CH2 amino acid sequence, and Domain K has a constant region domain amino acid sequence; (e) the fourth polypeptide chain comprises Domain L and Domain M; wherein the domains are arranged N-terminus to C-terminus in an L-M orientation; Domain L has a VH amino acid sequence and Domain M has a CH1 amino acid sequence; (f) the third and fourth polypeptide chains associate through interactions between the H domain and the L domain and between the I domain and the M domain; (g) the first and third polypeptide chains associate via an interaction between the D domain and the J domain and an interaction between the E domain and the K domain to form the antibody construct.
2. The antibody construct of claim 1 , wherein the amino acid sequences of the B domain and the G domain are identical, and the sequence is an endogenous CH3 amino acid sequence.
3. The antibody construct of claim 1 , wherein the B domain and the G domain contain a mutation that creates an engineered disulfide bridge between the B and G domains.
4. The antibody construct of claim 3, wherein the mutations generating engineered disulfide bridges are an S354C mutation in one of the B and G domains and a 349C mutation in the other domain.
5. The antibody construct of claim 1 , wherein the B domain and the G domain comprise knob-in-hole mutations.
6. 6. The antibody construct of claim 5, wherein the knob-in-hole mutation is a T366W mutation in one of the B and G domains, and T366S, L368A and Y407V mutations in the other domain.
7. The antibody construct of claim 1 , wherein the B domain and the G domain comprise charge pair mutations.
8. 8. The antibody construct of claim 7, wherein the charge pair mutation is a T366K mutation in one of the B and G domains and a L351D mutation in the other domain.
9. 9. The antibody construct of claim 1, wherein said domain E has a CH3 amino acid sequence.
10. (i) the amino acid sequences of the E and K domains are identical, and the sequences are endogenous CH3 amino acid sequences; or (ii) the amino acid sequences of the E and K domains are different; An antibody construct according to any one of claims 1 to 9.
11. 11. The antibody construct of claim 1, wherein the E and K domains comprise a mutation that creates an engineered disulfide bridge between domains E and K.
12. The antibody construct of claim 11, wherein the mutations generating engineered disulfide bridges are an S354C mutation in one of the E and K domains and a 349C mutation in the other domain.
13. The antibody construct of claim 1 , wherein the E domain and the K domain comprise charge pair mutations.
14. 14. The antibody construct of claim 13, wherein the charge pair mutation is a T366K mutation in one of the E or K domains and a corresponding L351D mutation in the other domain.
15. The antibody construct of any one of claims 1 to 10, wherein the amino acid sequences of the E domain and the K domain are endogenous sequences of two different antibody domains selected to have specific interactions that promote specific association between the first polypeptide chain and the third polypeptide chain.
16. The antibody construct of claim 15, wherein the two different amino acid sequences are a CH1 amino acid sequence and a CL amino acid sequence.
17. Domain E has a CH3 amino acid sequence; 15. The antibody construct of claim 1 , wherein domain K has a CH3 amino acid sequence.
18. 18. The antibody construct of any one of claims 1 to 17, wherein the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen, and the interaction between the H domain and the L domain forms a second antigen-binding site specific for a second antigen.
19. further comprising a fifth polypeptide chain, (a) the first polypeptide chain further comprises a Domain N and a Domain O, wherein the domains are arranged, from N-terminus to C-terminus, in the following orientation: NO-A-B-D-E; Domain N has a VL amino acid sequence and Domain O has a CH3 amino acid sequence; (b) the fifth polypeptide chain comprises a domain P and a domain Q; wherein the domains are arranged in a PQ orientation, from N-terminus to C-terminus, Domain P has a VH amino acid sequence and Domain Q has a CH3 amino acid sequence; and (c) the first and fifth polypeptide chains associate via interactions between the N domain and the P domain and between the O domain and the Q domain to form an antibody construct. The antibody construct of any one of claims 1 to 17.
20. (a) the amino acid sequences of domain N and domain A are identical; the amino acid sequence of domain H differs from the amino acid sequences of domains N and A; The amino acid sequences of domain O and domain B are identical; the amino acid sequence of domain I differs from the amino acid sequences of domains O and B; The amino acid sequences of domain P and domain F are identical; the amino acid sequence of domain L differs from the amino acid sequences of domains P and F; The amino acid sequences of domain Q and domain G are identical; the amino acid sequence of domain M differs from the amino acid sequences of domains Q and G; and (b) the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen, the interaction between the H domain and the L domain forms a second antigen-binding site specific for a second antigen, and the N and P domains form a third antigen-binding site specific for the first antigen.
21. (a) the amino acid sequences of domain N, domain A, and domain H are different; The amino acid sequences of domain O, domain B, and domain I are different; The amino acid sequences of domains P, F, and L are different; the amino acid sequences of Domain Q, Domain G, and Domain M are different; and (b) the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen; the interaction between the H domain and the L domain forms a second antigen-binding site specific for a second antigen; 20. The antibody construct of claim 19, wherein Domain N and Domain P form a third antigen-binding site specific for a third antigen.
22. further comprising a sixth polypeptide chain, (a) the third polypeptide chain further comprises a Domain R and a Domain S; wherein the domains are arranged, from N-terminus to C-terminus, in the orientation R-S-H-I-J-K; Domain R has a VL amino acid sequence and Domain S has a constant domain amino acid sequence; (b) the sixth polypeptide chain comprises a domain T and a domain U; wherein the domains are arranged in a TU orientation from N-terminus to C-terminus, Domain T has a VH amino acid sequence and Domain U has a constant domain amino acid sequence; and (c) the third and sixth polypeptides associate via interactions between the R and T domains and between the S and U domains to form the antibody construct.
23. (a) the amino acid sequences of domain R and domain A are identical; the amino acid sequence of domain H is different from the amino acid sequences of domains R and A; The amino acid sequences of domain S and domain B are identical; the amino acid sequence of domain I is different from the amino acid sequences of domains S and B; The amino acid sequences of domain T and domain F are identical; the amino acid sequence of domain L differs from the amino acid sequences of domains T and F; The amino acid sequences of domain U and domain G are identical; the amino acid sequence of domain M differs from the amino acid sequences of domains U and G; and (b) the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen; the interaction between the H domain and the L domain forms a second antigen-binding site specific for a second antigen; 23. The antibody construct of claim 22, wherein said Domain R and Domain T form a third antigen-binding site specific for said first antigen.
24. (a) the amino acid sequences of domain R and domain H are identical; the amino acid sequence of domain A differs from the amino acid sequences of domains R and H; The amino acid sequences of domain S and domain I are identical; the amino acid sequence of domain B differs from the amino acid sequences of domains S and I; The amino acid sequences of domain T and domain L are identical; the amino acid sequence of domain F differs from the amino acid sequences of domains T and L; The amino acid sequences of domain U and domain M are identical; the amino acid sequence of domain G differs from the amino acid sequences of domains U and M; and (b) the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen; the interaction between the H domain and the L domain forms a second antigen-binding site specific for a second antigen; 23. The antibody construct of claim 22, wherein the R and T domains form a third antigen-binding site specific for the second antigen.
25. (a) the amino acid sequences of domain R, domain A, and domain H are different; The amino acid sequences of domain S, domain B, and domain I are different; The amino acid sequences of domain T, domain F, and domain L are different; the amino acid sequences of domains U, G, and M are different; and (b) the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen; the interaction between the H domain and the L domain forms a second antigen-binding site specific for a second antigen; 23. The antibody construct of claim 22, wherein said Domain R and Domain T form a third antigen-binding site specific for a third antigen.
26. further comprising a fifth and a sixth polypeptide chain; (a) the first polypeptide chain further comprises a Domain N and a Domain O, wherein the domains are arranged from N-terminus to C-terminus in the following orientation: NO-A-B-D-E; (b) the third polypeptide chain further comprises Domain R and Domain S, wherein the domains are arranged from N-terminus to C-terminus in the following orientation: R-S-H-I-J-K; (c) the fifth polypeptide chain comprises a domain P and a domain Q, wherein the domains are arranged N-terminus to C-terminus in a PQ orientation; the sixth polypeptide chain comprises a domain T and a domain U, wherein the domains are arranged N-terminus to C-terminus in a TU orientation; and (d) the first and fifth polypeptide chains associate through an interaction between the N domain and the P domain and an interaction between the O domain and the Q domain; 18. The antibody construct of any one of claims 1 to 17, wherein the third and sixth polypeptide chains associate via interactions between the R and T domains and between the S and U domains to form the antibody construct.
27. (a) the amino acid sequences of domain N and domain A are identical; The amino acid sequences of domain H and domain R are identical; The amino acid sequences of domain O and domain B are identical; The amino acid sequences of domain I and domain S are identical; The amino acid sequences of domain P and domain F are identical; The amino acid sequences of domain L and domain T are identical; The amino acid sequences of domain Q and domain G are identical; the amino acid sequences of domain M and domain U are identical; and (b) the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen, the N domain and the P domain form a second antigen-binding site specific for the first antigen, the interaction between the H domain and the L domain forms a third antigen-binding site specific for a second antigen, and the interaction between the R domain and the T domain forms a fourth antigen-binding site specific for the second antigen.
28. (a) the amino acid sequences of domain H and domain A are identical; The amino acid sequences of domain N and domain R are identical; The amino acid sequences of domain I and domain B are identical; The amino acid sequences of domain O and domain S are identical; the amino acid sequences of domain L and domain F are identical; The amino acid sequences of domain P and domain T are identical; The amino acid sequences of domain M and domain G are identical; the amino acid sequences of Domain Q and Domain U are identical; and (b) the interaction between the A domain and the F domain forms a first antigen-binding site specific for a first antigen, the N domain and the P domain form a second antigen-binding site specific for a second antigen, the interaction between the H domain and the L domain forms a third antigen-binding site specific for the first antigen, and the interaction between the R domain and the T domain forms a fourth antigen-binding site specific for the second antigen.
29. 29. The antibody construct of any one of claims 1 to 28, wherein the sequence forming the junction between the A domain and the B domain is IKRTPREP or IKRTVREP.
30. 30. The antibody construct of any one of claims 1 to 29, wherein the sequence forming the junction between the F domain and the G domain is SSASPREP.
31. 31. The antibody construct of any one of claims 1 to 30, wherein at least one CH3 amino acid sequence has a C-terminal tripeptide insertion linking the CH3 amino acid sequence to a hinge amino acid sequence, wherein the tripeptide insertion is selected from the group consisting of PGK, KSC and GEC.
32. 32. The antibody construct of any one of claims 1 to 31, wherein at least one CH3 amino acid sequence is an IgG amino acid sequence.
33. 33. The antibody construct of claim 32, wherein the IgG amino acid sequence is an IgG1 amino acid sequence.
34. 34. An antibody construct according to any one of claims 1 to 33, wherein at least one CH3 amino acid sequence has one or more isoallotypic mutations.
35. The antibody construct of claim 34, wherein the isoallotypic mutations are D356E and L358M.
36. The CL amino acid sequence is C カッパ Amino acid sequence or C ラムダ 36. An antibody construct according to any one of claims 1 to 35, which is an amino acid sequence.
37. An antibody construct according to any one of claims 1 to 36, A pharma- ceutically acceptable carrier; 13. A pharmaceutical composition comprising:
38. 40. The pharmaceutical composition of claim 37 for use in a method of treatment, wherein said method of treatment comprises administering to a subject in need of treatment the pharmaceutical composition of claim 37.
39. 40. Use of the pharmaceutical composition of claim 37 in the manufacture of a medicament.
Citation Information
Patent Citations
Method and means for producing Ig-like molecules
JP2015515275A
Tandem Fc bispecific antibody
JP2015527869A
JPP726115B
Multispecific antibodies
US20150315296A1
Domain-exchanged antibody
WO2016087650A1
