Stabilizing homodimer mutations for two cell heterodimer production

Abstract

The present invention relates to methods of producing homodimers comprising mutations in the CH3 domains. Said mutations increase the stability of the homodimer, resulting in increased homodimer yield. Said homodimers are separately expressed and purified, and reassembled into heterodimers.

Description

STABILIZING HOMODIMER MUTATIONS FOR TWO CELLHETERODIMER PRODUCTIONFIELD OF THE INVENTION

[0001] The present invention relates to methods of producing homodimers comprising mutations in the CH3 domains. Said mutations increase the stability of the homodimer, resulting in increased homodimer yield. Said homodimers are separately expressed and purified, and reassembled into heterodimers.DESCRIPTION OF THE TEXT FILE SUBMITTED ELECTRONICALLY

[0002] The present application contains a Sequence Listing, which has been submitted electronically in XML format and is hereby incorporated by reference in its entirety. The computer readable format copy of the Sequence Listing, which was created on December 11, 2024, is named 10284-W001-SEC_ST26.xml and is 12,591 bytes in size.BACKGROUND OF THE INVENTION

[0003] Heterodimers can be difficult to produce at high purities and yields due to chain mispairing. For example, two antibodies that are co-expressed can lead to a mixture of possible antibody products in addition to the desired heterodimer (see e.g., United States Patent Number 9,150,663). A two-cell approach for producing heterodimers remedies the light chain pairing problem, however, low expression levels of each homodimer in the two-cell approach still limits the total yield of the final assembled heterodimer product.

[0004] Strategies to facilitate correct pairing of two heavy chains in the heterodimer include the knobs-into-holes approach (see e.g. Ridgway et al., Protein Engineering vol.9 no.7 pp.617-621, 1996) and the charge pair mutation approach. Earlier work on driving heterodimeric chain pairing between IgG heavy chains began with the use of the knob-into- holes strategy in which the contours of the surface area between chains are modified with mutations to larger and smaller residues. A more recent approach that has been implemented broadly is the use of charge pair mutations, in which naturally charged residues at one of thetwo rotationally symmetrical regions are reversed by mutating the positively charged residue to a negatively charged residue and vice versa (PCT Publication Number WO 2021 / 092355).

[0005] The present invention describes additional engineered mutations nearby charge pair mutation networks that facilitate heavy chain pairing that were able to rescue poor homodimer expression and greatly improve final yields of assembled heterodimer products, while having little effect on assembly efficiencies and thermal stabilities of the heterodimers.SUMMARY OF THE INVENTION

[0006] The present invention provides a method for producing a heterodimer, wherein the method comprises expressing a first homodimer in a first cell, and expressing a second homodimer in a second cell; purifying the first and second homodimers, wherein the first and second homodimers are combined together either before purifying or after purifying; subjecting the combined first and second homodimers to reducing conditions; subjecting the combined first and second homodimers to oxidizing conditions; and obtaining the heterodimer; wherein the first homodimer comprises different mutations compared to the second homodimer, and wherein the mutations are according to EU numbering and are selected from the group consisting of:

[0007] E356K and D399K;

[0008] K392D, K409D, and K439D;

[0009] E356K, D399K, and one of K370E, K370L, K370M, K370Q, and K370T;

[0010] K409R and one of D399 A, D3991, D399L, D399M, D399N, D399P, D399Q,D399S, D399T, and D399V;

[0011] one of K409A, K409I, K409L, K409M, K409N, K409P, K409Q, K409S,K409T, and K409V;

[0012] K409R and one of E356A, E356I, E356L, E356M, E356N, E356P, E356Q,E356S, E356T, and E356V;

[0013] K409R and one of K439 A, K4391, K439L, K439M, K439N, K439P, K439Q,K439S, K439T, and K439V;

[0014] D399K;

[0015] K392D and K409D;

[0016] K439D and D399K; and

[0017] E356K, K392D, and K409D.

[0018] In an embodiment, the first homodimer is not expressed in the second cell, and the second homodimer is not expressed in the first cell. In an embodiment, the first and second homodimers are combined together before purifying the first and second homodimers. In an embodiment, the first and second homodimers are combined after separately purifying each of the first and second homodimers.

[0019] The present invention provides a method of increasing the yield of a homodimer, wherein the method comprises expressing a homodimer comprising mutation(s) of the present invention.

[0020] The present invention provides a method of increasing the yield of a homodimer, wherein the method comprises expressing a first homodimer in a first cell, and expressing a second homodimer in a second cell; purifying the first and second homodimers, wherein the first and second homodimers are combined together either before purifying or after purifying; subjecting the combined first and second homodimers to reducing conditions; subjecting the combined first and second homodimers to oxidizing conditions; and obtaining a heterodimer; wherein the first homodimer comprises different mutations compared to the second homodimer, and wherein the mutations are according to EU numbering and are selected from the group consisting of

[0021] E356K and D399K;

[0022] K392D, K409D, and K439D;

[0023] E356K, D399K, and one of K370E, K370L, K370M, K370Q, and K370T;

[0024] K409R and one of D399 A, D3991, D399L, D399M, D399N, D399P, D399Q,D399S, D399T, and D399V;

[0025] one of K409A, K409I, K409L, K409M, K409N, K409P, K409Q, K409S,K409T, and K409V;

[0026] K409R and one of E356A, E356I, E356L, E356M, E356N, E356P, E356Q,E356S, E356T, and E356V;

[0027] K409R and one of K439 A, K4391, K439L, K439M, K439N, K439P, K439Q,K439S, K439T, and K439V;

[0028] D399K;

[0029] K392D and K409D;

[0030] K439D and D399K; and

[0031] E356K, K392D, and K409D.

[0032] In an embodiment, the first homodimer is not expressed in the second cell, and the second homodimer is not expressed in the first cell. In an embodiment, the first and second homodimers are combined before purifying the first and second homodimers. In an embodiment, the first and second homodimers are combined after separately purifying each of the first and second homodimers.

[0033] The present invention provides a homodimer, wherein the homodimer comprises mutations according to EU numbering selected from the group consisting of

[0034] E356K and D399K;

[0035] K392D, K409D, and K439D;

[0036] E356K, D399K, and one of K370E, K370L, K370M, K370Q, and K370T;

[0037] K409R and one of D399 A, D3991, D399L, D399M, D399N, D399P, D399Q,D399S, D399T, and D399V;

[0038] one of K409A, K409I, K409L, K409M, K409N, K409P, K409Q, K409S, K409T, and K409V;

[0039] K409R and one of E356A, E356I, E356L, E356M, E356N, E356P, E356Q,E356S, E356T, and E356V;

[0040] K409R and one of K439 A, K4391, K439L, K439M, K439N, K439P, K439Q,K439S, K439T, and K439V;

[0041] D399K;

[0042] K392D and K409D;

[0043] K439D and D399K; and

[0044] E356K, K392D, and K409D.

[0045] The present invention provides a heterodimer, wherein the heterodimer comprises the following mutations according to EU numbering:

[0046] E356K and D399K and one of K370E, K370L, K370M, K370Q, and K370T in one half of the heterodimer, and K392D, K409D, and K439D in the other half of the heterodimer;

[0047] K409R and one of D399 A, D3991, D399L, D399M, D399N, D399P, D399Q,D399S, D399T, and D399V in one half of the heterodimer; and one of K409A, K409I, K409L, K409M, K409N, K409P, K409Q, K409S, K409T, and K409V in the other half of the heterodimer;

[0048] K409R and one of E356A, E356I, E356L, E356M, E356N, E356P, E356Q,E356S, E356T, and E356V in one half of the heterodimer; and K409R and one of K439A,K439I, K439L, K439M, K439N, K439P, K439Q, K439S, K439T, and K439V in the other half of the heterodimer;

[0049] D399K in one half of the heterodimer; and K392D and K409D in the other half of the heterodimer; or

[0050] K439D and D399K in one half of the heterodimer; and E356K, K392D, andK409D in the other half of the heterodimer.

[0051] In an embodiment, the heterodimer comprises one half of the first homodimer(first homodimer half) and one half of the second homodimer (second homodimer half), and wherein the heterodimer comprises the following mutations according to EU numbering:

[0052] E356K and D399K and one of K370E, K370L, K370M, K370Q, and K370T in the first homodimer half, and K392D, K409D, and K439D in the second homodimer half;

[0053] K409R and one of D399 A, D3991, D399L, D399M, D399N, D399P, D399Q,D399S, D399T, and D399V in the first homodimer half; and one of K409A, K409I, K409L, K409M, K409N, K409P, K409Q, K409S, K409T, and K409V in the second homodimer half;

[0054] K409R and one of E356A, E356I, E356L, E356M, E356N, E356P, E356Q,E356S, E356T, and E356V in the first homodimer half; and K409R and one of K439A, K439I, K439L, K439M, K439N, K439P, K439Q, K439S, K439T, and K439V in the second homodimer half;

[0055] D399K in the first homodimer half; and K392D and K409D in the second homodimer half; or

[0056] K439D and D399K in the first homodimer half; and E356K, K392D, andK409D in the second homodimer half.

[0057] The present invention provides a homodimer of the present invention for use in the treatment of a disease.

[0058] The present invention provides a heterodimer of the present invention for use in the treatment of a disease.

[0059] The present invention provides a homodimer of the present invention for use in therapy.

[0060] The present invention provides a heterodimer of the present invention for use in therapy.

[0061] The present invention provides a homodimer of the present invention for the manufacture of a medicament for the treatment of a disease.

[0062] The present invention provides a heterodimer of the present invention for the manufacture of a medicament for the treatment of a disease.

[0063] In an embodiment, the first or second homodimer, or first or second homodimer half of the heterodimer, comprises E356M and K409R mutations.

[0064] In an embodiment, the first or second homodimer, or first or second homodimer half, comprises K439Q and K409R mutations.

[0065] In an embodiment, the first or second homodimer, or first or second homodimer half, comprises E356T and K409R mutations.

[0066] In an embodiment, the first or second homodimer, or first or second homodimer half, comprises E356K, D399K, and K370E mutations.

[0067] In an embodiment, the first or second homodimer, or first or second homodimer half, comprises K439Q and K409R mutations.

[0068] In an embodiment, the first or second homodimer, or first or second homodimer half, comprises E356N and K409R mutations.

[0069] In an embodiment, the first or second homodimer, or first or second homodimer half, comprises E356T and K409R mutations.

[0070] In an embodiment, the first or second homodimer, or first or second homodimer half, comprises K439T and K409R mutations.

[0071] In an embodiment, the heterodimer comprises E356K, D399K, and K370E mutations in the first homodimer half; and K392D, K409D, and K439D mutations in the second homodimer half.

[0072] In an embodiment, the heterodimer comprises E356N and K409R mutations in the first homodimer half; and K439Q and K409R mutations in the second homodimer half.

[0073] In an embodiment, the heterodimer comprises E356T and K409R mutations in the first homodimer half; and K439Q and K409R mutations in the second homodimer half.

[0074] In an embodiment, the heterodimer comprises E356T and K409R mutations in the first homodimer half; and K439T and K409R mutations in the second homodimer half.

[0075] In an embodiment, the first homodimer and / or the second homodimer are produced at a higher yield compared to a homodimer comprising only E356K and D399K mutations or compared to a homodimer comprising only K392D, K409D, and K439D mutations.

[0076] The present invention provides a heterodimer produced according to a method of the present invention.

[0077] The present invention provides a pharmaceutical composition comprising a heterodimer of the present invention.

[0078] In an embodiment, said mutations are in the HC CH3 domain. In an embodiment, the mutations in the HCs promote pairing of the HCs with one another.

[0079] In an embodiment, heterodimer assembly is facilitated by charge pair mutations.

[0080] In an embodiment, the reducing conditions comprise mixing homodimers with about 5 mM GSH, resulting in a reduced mixture. In an embodiment, the reduced mixture is incubated at about 36°Cfor about five hours. In an embodiment, after incubating for about five hours, the reduced mixture is buffer exchanged into IX PBS, pH 7.4 and incubated for 12-16 hours at about 36°C.

[0081] In an embodiment, the reducing conditions comprise mixing homodimers with about 25 mM 2-MEA, resulting in a reduced mixture. In an embodiment, the reduced mixture is incubated at about 22°C for at least 5 hours. In an embodiment, the reduced mixture is incubated at about 22°C for at least 12 hours. In an embodiment, after incubating, the samples are exchanged into Tris pH 7.4 to a final concentration of about 100 mM and incubated at about 22°C for 12-16 hours.

[0082] In an embodiment, a heterodimer of the present invention binds two different targets. In an embodiment, a heterodimer of the present invention binds a tumor target. In an embodiment, a heterodimer of the present invention binds two different tumor targets.BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 depicts chain pairing challenges in heterodimers. Heterodimers are comprised of 4 unique chains, which often mispair due to non-specific interactions, leading to diminished productivity of correctly paired heterodimers.

[0084] Figure 2 depicts a two-cell platform for production of heterodimers. Two halves of a heterodimer are independently expressed in two separate cell cultures as homodimers such that the cognate HCs and LCs are correctly paired. After purification of the KK and DDD homodimers, the purified homodimers are mixed under reducing conditions to separate the HCs into / i antibodies. Next, under oxidizing conditions, the correct HCs of the heterodimer are reassembled, facilitated by coordinated charge pair mutations (CPMs) in the Fc (black and white dots) between the two unique HCs.

[0085] Figures 3A and 3B depict clustering of charged residues in KK and DDD homodimers that might contribute to poor expression. Figure 3a) Expression titers of KK and DDD homodimers across 3 Target 1 binders and 3 Target 2 binders. Figure 3b) Models of IgGl CH3 domains in a KK homodimer (left) and a DDD homodimer (right). All residues that participate in CPM networks are shown as sticks, as well as a neighboring residue, K370. Residues that participate in the CPM networks in the heterodimer are boxed, or shown faintly for equivalent networks on the opposite side of the interface (labeled ’).

[0086] Figure 4 depicts design strategies to improve stability and yield of homodimers. KK homodimers comprise mutations E356K and D399K, while DDD homodimers comprise mutations K392D, K409D, and K439D. KK repacked homodimers comprise mutations E356K, D399K, and K370E / L / M / Q / T. Alternative interaction homodimers comprise mutations K409R and D399A / I / L / M / N / P / Q / S / T / V in one homodimer, while the other homodimer comprises mutations K409A / I / L / M / N / P / Q / S / T / V. Another set of alternative interaction homodimers comprise mutations K409R and E356A / I / L / M / N / P / Q / S / T / V in one homodimer, while the other homodimer comprises mutations K409R AND K439A / I / L / M / N / P / Q / S / T / V. Single CPM homodimers comprise a D399K mutation in one homodimer, while the other homodimer comprises mutations K392D and K409D. Mixed CPM homodimers comprise mutations K439D and D399K in one homodimer, while the other homodimer comprises mutations E356K, K392D, and K409D.

[0087] Figures 5A and 5B depict titers of KK and DDD homodimers. Figure 5a) Titers of the anti-Target 1 KK homodimer controls and variants are shown as percentages compared to the WT control with no CPMs. Figure 5b) Titers of the anti-Target 2 DDD homodimer control and variants are shown as percentages compared to the WT control with no CPMs.

[0088] Figures 6A and 6B depict purity of assembled heterodimers. Figure 6a) % purities of correctly assembled heterodimers for the KK / DDD control and the mixed CPM, single CPM, and KK repacked variants as determined by nrMCE. Figure 6b) %MCE purities of correctly assembled heterodimers for alternative interaction variants. The %MCE purities of the KK / DDD controls are indicated by the top, left-most cell. Cells are colored by binned %purities as indicated by the scale on the right. Striped cells indicate where no data was attainable.

[0089] Figure 7 depicts the structural comparison of CH3 heterodimers with KK / DDD CPMs with and without K370E. The K392D / K409D / D399K CPM network is shown on the left(PDB 71us), and a model of the same CPM network with the K370E mutation is showed on the right.

[0090] Figure 8 depicts the thermostability of assembled heterodimers.

[0091] Figure 9 depicts homodimer yields and percent purities of assembled target3 / target 4 heterodimers produced at IL scale.DETAILED DESCRIPTION

[0092] Heterodimers (e.g. heterodimeric antibodies) comprise a significant portion of current drug development efforts but remain challenging to produce due to the tendency of their four unique antibody chains to mispair (Figure 1). While reliable and platform-wide engineering solutions exist to correctly pair the two heavy chains (HCs) of a heterodimer, solutions to correctly pair each HC with its cognate light chain (LC) work with varying degrees of success and often need to be readapted for each heterodimer pair due to the uniqueness of the variable heavy (VH) and variable light (VL) domains for each binder and the diversity in LCs (i.e. lambda vs. kappa LCs).

[0093] One solution for producing heterodimers with correct HC / LC pairing is to first pair the cognate HCs and LCs independently by expressing the two halves of the heterodimer in two separate cell cultures, from which each half can be purified separately as homodimers (Figure 2). The homodimers are then mixed, and the HCs of the homodimers are separated and re-assembled into heterodimers in a redox reaction, where correct HC / HC pairing is facilitated by coordinated charge pair mutations (CPMs) in the Fc region. Despite significantly improved HC / LC pairing using this two-cell approach, poor expression of the homodimers still results in generally low yield of the assembled heterodimers using the two-cell method (Figure 3a).

[0094] As described herein, it was observed that homodimers containing positively charged CPMs, termed KK (E356K and D399K) homodimers, have significantly reduced titers across three different Target 1 binders compared to titers for DDD (K392D, K409D, and K439D) homodimers across three Target 2 binders. Therefore the consistently reduced titers of KK homodimers contribute more significantly to reduced total yields of assembled heterodimers.

[0095] The CPMs used to drive correct HC pairing in heterodimers form two main saltbridge networks in the heterodimer: 1) D399K from the KK homodimer and K392D and K409D from the DDD homodimer, and 2) E356K from the KK homodimer and K439D from the DDD homodimer. As a solution to the low productivity of KK and DDD homodimers,mutations were designed to stabilize the homodimers and whether these mutations could rescue low titer and yield while maintaining efficient heterodimer assembly and stability of the assembled products was examined.

[0096] The positions of amino acids are given based on EU numbering. It is understood that if there is a different wild-type residue at a position than what is stated herein, that position can still be mutated to the given residue. For example, if a HC comprises D356 instead of E356, it is envisioned that either amino acid at position 356 can be mutated to lysine (K). Unless otherwise noted herein, the wild-type residues are given based on a wild-type residue found in IgGl. It is understood that an IgGl CH3 might have different allotypes variations or other variations and / or modifications, and all such variants are encompassed within the present invention, provided they have at least one mutation described herein. IgGl CH3 comprising at least one mutation of the present invention and the following mutations are also included: R292C, N297G, V302C.

[0097] By way of example, D356 in IgGl is E356 in IgG2, IgG3, and IgG4. D399 is D399 in all IgGl, IgG2, IgG3, and IgG4. K392 is K392 in IgGl, IgG2, and IgG4, and is N392 in IgG3. K409 is K409 in IgGl, IgG2, and IgG3, and is R409 in IgG4. K439 is K439 in all IgGl, IgG2, IgG3, and IgG4. Separate expression of two parental antibodies that are subsequently each purified, separated and reassembled to form heterodimers is disclosed herein and in Labrijn et. al, PNAS Vol. 110, No. 13 (2013). Said methods are contemplated as part of the present invention for separate expression of homodimers, and subsequent purification and reassembly into heterodimers.

[0098] By way of example, a human germline IgGl CH3 sequence translated from J00228 is given by SEQ ID NO: 1, and includes a C-terminal lysine. It is understood that when expressed, the cell will normally clip the C-terminal lysine. There are 4 positions in the sequence that have observed polymorphisms.

[0099] A person of ordinary skill can readily determine the positions of the mutations of the present invention by EU numbering. However, by way of example, nonlimiting examples of sequences comprising mutations of the present invention include:

[0100] IgGl CH3 (SEQ ID NO: 1 and SEQ ID NO: 2), and

[0101] IgGl CH3 comprising:K439Q and K409R mutations (SEQ ID NO: 3), K439T and K409R mutations (SEQ ID NO: 4), K439Q and K409R mutations (SEQ ID NO: 5),K439L and K409R mutations (SEQ ID NO: 6), K392D, K409D and K439D mutations (SEQ ID NO: 7), E356T and K409R mutations (SEQ ID NO: 8), E356N and K409R mutations (SEQ ID NO: 9), E356I and K409R mutations (SEQ ID NO: 10),E356K, D399K and K370E mutations (SEQ ID NO: 11), and E356K and D399K mutations (SEQ ID NO: 12).

[0102] These mutations are in the context of SEQ ID NO: 2, but they may also be in the context of SEQ ID NO: 1 or other CH3 variants.

[0103] Homodimers can be produced by cotransfecting correctly paired HC and LC expression vectors in a CHO cell line using Lipofectamine LTX (Gibco). Antibodies are then purified by Protein A affinity chromatography (Mab Select SuRe), and the samples neutralized to pH 8. To assemble heterodimers, equimolar amounts of homodimers are mixed and incubated with reduced glutathione (GSH) at a final concentration of 5 mM for 5 hours at 36°C. The mixtures are buffer exchanged into IX PBS, pH7.4 to remove GSH and incubated for 12- 16 hours at 36°C.

[0104] In another method to assemble heterodimers, equimolar amounts of homodimers can be mixed and incubated with 2-Mercaptoethylamine (2-MEA) at a final concentration of 25 mM for 12 hours at 22°C. The mixtures are then desalted into Tris pH 7.4 to a final concentration of 100 mM and incubated for 12-16 hours at 22°C.

[0105] 2-Mercaptoethylamine.HCl (2-MEA) is a mild reducing agent that has been described to selectively cleave disulfide bonds in the hinge region of antibodies, while preserving the disulfide bonds between the heavy and light chains.

[0106] “About” refers to 10% + / - a stated value.

[0107] A "homodimer" refers to a paired antibody HC and LC that is paired with an identical paired antibody HC and LC. Each antibody HCs comprises the same mutation(s) as the other HC. In an embodiment, said mutations are in the HC CH3 domain.

[0108] A "heterodimer" refers to an antibody HC that is paired with a different antibody HC. Each antibody HC comprises different mutation(s) compared to the other antibody HC. Each HC is paired with a LC, and preferably, each LC is different from the other LC. However, it is also contemplated that each HC is paired with a LC and each LC is the same as the other LC in the context of heterodimer with two different antigen binding HCs that pair to a common light chain (see e.g., PCT Publication Number WO 2016 / 079081 for a description of a commonlight chain). Thus, a heterodimer comprises one half of two non-identical homodimers. In an embodiment, said mutations are in the HC CH3 domain. In an embodiment, the mutations in the HCs promote pairing of the HCs with one another.

[0109] As used herein, “KK repacked” refers to a monomer comprising E356K, D399K, and K370E / L / M / Q / T mutations. In the context of a heterodimer, KK repacked refers to E356K, D399K, and K370E / L / M / Q / T mutations in one half of the heterodimer, and K392D, K409D, and K439D mutations in the other half of the heterodimer.

[0110] Alternative interaction heterodimer variants comprised either: 1) 2 mutations including D399A / I / L / M / N / P / Q / S / T / V and K409R on one half of the heterodimer and one of K409A / I / L / M / N / P / Q / S / T / V mutation on the other half, or 2) 2 mutations including E356A / I / L / M / N / P / Q / S / T / V and K409R on one half of the heterodimer and 2 mutations including K439A / I / L / M / N / P / Q / S / T / V and K409R on the other half.

[0111] Charge pair mutations used for heterodimer assembly are described in PCT Publication Number WO 2009 / 089004. In general, one or more residues that make up the CH3- CH3 interface is replaced with a charged amino acid such that the interaction becomes electrostatically unfavorable in the homodimer and electrostatically favorable in the heterodimer. This strategy alters a CH3 domain to reduce the ability of the domain to interact with itself, i.e., form homodimers, and to favor the ability of the domain to interact with an oppositely charged CH3 domain, i.e. form heterodimers.

[0112] KK homodimer refers to E356K and D399K charge pair mutations.

[0113] DDD homodimer refers to K392D, K409D, and K439D charge pair mutations.

[0114] It is also envisioned that a heterodimer of the present invention may be attached to one or more moieties such as scFv, peptide, protein, small molecule, and / or drug. In an embodiment, said moiety is attached to the C-terminal region of the heterodimer HC. In an embodiment, said moiety is attached to the C-terminal region of the heterodimer LC. In an embodiment, said moiety is attached to the N-terminal region of the heterodimer HC. In an embodiment, said moiety is attached to the N-terminal region of the heterodimer LC. In the case of protein attachments, these can be antibody-protein fusions (e.g. antibody-cytokine fusions) or different multispecific antibody formats (e.g. adding another scFv binder). It is envisioned that a heterodimer of the present invention can be used as an antibody-drug conjugate. Attachment may occur by methods known in the art (see e.g., McCombs et. al, AAPS J. 2015 Mar; 17(2): 339-351; and Sawant et. al, Int J Mol Sci. 2020 Oct; 21(20): 7496).Peptides and small molecules / drugs can be attached throughout antibody domains via a chemical linker - these are called ADCs.

[0115] As used herein, an “antibody” is an immunoglobulin molecule comprising 2 heavy chains (HCs) and 2 light chains (LCs) interconnected by disulfide bonds. The amino terminal portion of each LC and HC includes a variable region of about 100-120 amino acids primarily responsible for antigen recognition via the CDRs contained therein. The CDRs are separated with regions that are more conserved, termed framework regions (“FR”). Each LCVR and HCVR is composed of 3 CDRs and 4 FRs, arranged from amino-terminus to carboxyterminus in the following order: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4. The 3 CDRs of the LC are referred to as “LCDR1, LCDR2, and LCDR3,” and the 3 CDRs of the HC are referred to as “HCDR1, HCDR2, and HCDR3.” The CDRs contain most of the residues which form specific interactions with the antigen. The functional ability of an antibody to bind a particular antigen is, thus, largely influenced by the amino acid residues within the six CDRs. Assignment of amino acids to CDR domains within the LCVR and HCVR regions of the antibodies of the present invention is based on the well-known Kabat numbering convention (Kabat, et al., Ann. NY Acad. Sci. 190:382-93 (1971); Kabat et al., Sequences of Proteins of Immunological Interest, Fifth Edition, U.S. Department of Health and Human Services, NIH Publication No. 91-3242 (1991)). It is understand that other numbering conventions may also be used, such as, for example, Chothia (Chothia et al., “Canonical structures for the hypervariable regions of immunoglobulins”, Journal of Molecular Biology, 196, 901-917 (1987); Al-Lazikani et al., “Standard conformations for the canonical structures of immunoglobulins”, Journal of Molecular Biology, 273, 927-948 (1997)), and / or North (North et al., “A New Clustering of Antibody CDR Loop Conformations”, Journal of Molecular Biology, 406, 228-256 (2011)). The fragment crystallizable region (“Fc region”) is the end of the antibody HC that binds Fc receptors and proteins of the complement system. The IgG Fc region, for example, comprises two CH2 and two CH3 domains. In an embodiment, the heterodimer comprises one antibody arm (HC and LC) that binds one antigen, and another antibody arm (HC and LC) that binds a different antigen.

[0116] An "antigen" refers to a target molecule to which the homodimer or heterodimer binds.

[0117] Yield can be determined by methods known in the art. For example, protein concentrations can be determined by measuring the optical density (OD) at 280 nm using the molar extinction coefficient calculated from the protein's amino acid sequence. Protein yieldscan be calculated as the mass of protein yielded from purification divided by the load volume, which is the volume of a protein sample applied to a chromatography column during protein purification.

[0118] Non-limiting reducing and oxidizing conditions include mixing and incubating homodimers at a final concentration of 5 mM GSH at 36°C for 5 hours, and then buffer exchanging into IX PBS, pH 7.4 to remove GSH, and incubating for 12-16 hours at 36°C. Another method includes mixing and incubating homodimers at a final concentration of 25 mM 2-MEA at 22°C for at least 5 hours, and then desalting the mixture by adding Tris pH 7.4 to a final concentration of 100 mM and incubating at 22°C for 12-16 hours.

[0119] As used interchangeably herein, “treatment” and / or “treating” and / or “treat” are intended to refer to all processes wherein there may be a slowing, interrupting, arresting, controlling, stopping, or reversing of the progression of the disorders described herein, but does not necessarily indicate a total elimination of all disorder symptoms. Treatment includes administration of a homodimer or heterodimer of the present invention for treatment of a disease or condition in a human that would benefit from activity of a homodimer or heterodimer of the present invention, and includes: (a) inhibiting further progression of the disease; and (b) relieving the disease, i.e., causing regression of the disease or disorder or alleviating symptoms or complications thereof.

[0120] A homodimer or heterodimer of the present invention, or a pharmaceutical composition comprising the same, may be administered by parenteral routes, non-limiting examples of which are subcutaneous administration and intravenous administration. Intramuscular, intraarterial, intralesional, and peritoneal bolus injection are other possible routes of administration. A homodimer or heterodimer can also be administered via infusion, for example intravenous or subcutaneous infusion. A homodimer or heterodimer of the present invention may be administered to a patient with pharmaceutically acceptable carriers, diluents, or excipients in single or multiple doses. Optionally, the composition additionally comprises one or more physiologically active agents. Pharmaceutical compositions of the present invention can be prepared by methods well known in the art (e.g., Remington: The Science and Practice of Pharmacy, 22nd ed. (2012), A. Loyd et al., Pharmaceutical Press) and comprise a heterodimer or homodimer, as disclosed herein, and one or more pharmaceutically acceptable carriers, diluents, or excipients.EXAMPLESEXAMPLE 1: HETERODIMER ENGINEERING

[0121] Poor expression of homodimers may be due, at least in part, to the homodimers comprising multiple clusters of same-charge residues in close structural proximity (Figure 3b). In addition, in KK homodimers, an additional neighboring lysine side chain (K370) is predicted to lie within 4A of the D399K CPM; therefore mutations were designed at a K370 to better structurally accommodate the D399K mutation in the KK homodimers (Figure 4, KK repacked mutations).

[0122] In another set of mutations, either of the two salt-bridge networks in the assembled heterodimer were converted into hydrophobic or polar interactions such that: 1) E356K in the KK homodimer and K439D in the DDD homodimer, or 2) D399K in the KK homodimer and K409D in the DDD homodimer, were replaced by hydrophobic or polar residues including alanine, isoleucine, leucine, methionine, asparagine, proline, glutamine, serine, threonine, or valine (Figure 4, alternative interaction mutations; (hydrophobic or polar interactions are replacing the charged interactions in the charge pair mutations)). A K409R mutation (see e.g., United States Patent Number 9,150,663) was also added for the alternative interaction variants in both homodimers when possible to weaken the CH3 interaction and facilitate separation of the homodimer into U antibodies. Therefore alternative interaction heterodimer variants comprised either: 1) 2 mutations including D399A / I / L / M / N / P / Q / S / T / V and K409R on one half of the heterodimer and one of K409 A / I / L / M / N / P / Q / S / T / V mutation on the other half, or 2) 2 mutations including E356 A / I / L / M / N / P / Q / S / T / V and K409R on one half of the heterodimer and 2 mutations including K439 A / I / L / M / N / P / Q / S / T / V and K409R on the other half.

[0123] Upon redox assembly of the alternative interaction variants, all possible pairs of hydrophobic and polar residues were combinatorially explored in the assembled heterodimer products. In addition to alternative interaction variants, designs where either one of the two CPM networks were implemented independently in the assembled heterodimer (single CPM) and where the CPM residues were swapped between the KK and DDD homodimers (mixed CPM) were also engineered. Overall, 33 mutations in KK homodimers and 23 associatedmutations in DDD homodimers were engineered (non-limiting exemplary mutations are shown in Figure 4).

[0124] All homodimers were produced by cotransfecting correctly paired HC and LC expression vectors in a CHO cell line using Lipofectamine LTX (Gibco). Homodimers were purified by Protein A affinity chromatography (Mab Select SuRe), and the samples neutralized to pH 8. To assemble heterodimers, equimolar amounts of homodimers were mixed and incubated with reduced glutathione (GSH) at a final concentration of 5 mM for 5 hours at 36°C. The mixtures were buffer exchanged into IX PBS, pH7.4 to remove GSH and incubated for 12- 16 hours at 36°C. In another method to assemble heterodimers, equimolar amounts of homodimers were mixed and incubated with 2-Mercaptoethylamine (2-MEA) at a final concentration of 25 mM for 12 hours at 22°C. The mixtures were then desalted by adding Tris pH 7.4 to a final concentration of 100 mM and incubated for 12-16 hours at 22°C.EXAMPLE 2: ENGINEERED HOMODIMER TITERS

[0125] Mutations on KK homodimers were tested using two different Target 1 binders (Target 1 binder 1 and Target 1 binder 2) and on one anti-Target 2 DDD homodimer. Each LC and HC chain of the homodimers were cloned separately into pBMVl. l mammalian expression vectors via Golden Gate assembly of DNA fragments synthesized by Twist Bioscience. HCs were constructed in IgGl with SEFL2.2 (R309C / N314G / V321C) and YTE (M265Y / S267T / T269E) mutations in the CH2 domains, along with CPMs in the CH3 domains (E356K / D399K for KK homodimers and K392D / K409D / K439D for DDD homodimers). After sequence confirmation, DNA was prepared using Maxi plasmid purification kits, and plasmids were mixed at a mass-based ration of 1 : 1 : (LC:HC). The KK and DDD homodimers were stably expressed separately in in a CHO cell line expression hosts at 40 mL scale after transfection via Lipofectamine LTX (Gibco #15338-100). After selection and scaling of cultures, cultured media was harvested after 20 days of production, and titers were measured via Pro A capture.Of the variants tested, 28 out of 33 variants improved titers of KK homodimers compared to the WT KK homodimer controls for both Target 1 binders (Figure 5). 11 of the 33 KK homodimer variants showed improved titer even compared to the WT control in the KK homodimer using Target 1 binder 2, and one of these variants (E356T alternative interaction) showed improved titer compared to the WT control in both anti-Target 1 KK homodimers. This demonstrates that poor yield of the KK homodimer can be significantly improved by nearly 5-fold, and therefore yield of the assembled heterodimer will also be greatly improved. Titers for the anti-Target 2 DDD homodimers were also improved by nearly 2-fold for most DDD homodimer variants, which will also improve yields of assembled heterodimers.EXAMPLE 3: HETERODIMER ASSEMBLY

[0126] The KK and DDD homodimers were purified from the cultured media, then redox assembled for analysis of the assembled heterodimer products. Protein A affinity purification was used to capture the homodimers (Mab Select SuRe), then samples neutralized to pH 6.5 using 2M Tris, pH 8. Redox assembly of the KK and DDD homodimers into the final heterodimer products was performed by mixing equal parts of the appropriate KK and DDD homodimer and incubating for 5 hours at a final concentration of 5 mM reduced glutathione (GSH) at 36°C, followed by buffer exchange into IX PBS, pH 7.4 to remove GSH and incubation for 12-16 hours at 36°C. The purity of the correctly assembled heterodimers were then analyzed by non-reducing microfluidic capillary electrophoresis (MCE).

[0127] As shown in Figure 6, the assembled KK / DDD heterodimer controls assembled efficiently, with 74% or 80% of redox-assembled material representing the correctly assembled heterodimers containing either Target 1 binder 1 or Target 1 binder 2, respectively. The mixed CPM variants assembled less efficiently than the KK / DDD controls (47-53%), whereas the K / DD single CPM (D399K / K392D / K409D) and KD single CPM (E356K / K439D) assembled with similar efficiency (72-86%) as the KK / DDD control with the exception of the Target 1 binder 2-containing K / DD single CPM variant, which had only 11% of correctly paired product. The KK repacked variants assembled slightly more efficiently (75-82%) compared to the KK / DDD controls. Unexpectedly, the K370E KK repacked variant did not negatively affect heterodimerization, despite the loss of a potential salt bridge between K370 and K409D and introducing a potentially repulsive interaction between K370E and K409D. A structural model of the K370E mutation in the heterodimer predicts that the heavy atoms of K370 and K409D in the heterodimer are located 3.5-4.2A apart (Figure 7) and may therefore be unable to form a strong salt bridge, as reflected in the lack of preference for K370 in forming the assembled heterodimer (Figure 6a). However, the preference for the K370E mutation in the KK homodimer is clearly reflected in the ~5X increased titer of the homodimer upon adding the mutation (Figure 5a). Therefore the K370E mutation improves two-cell production of heterodimers by stabilizing the KK homodimer and increasing KK homodimer titer, rather than influencing the assembly of the heterodimer via CPM networks.In addition to the KK repacked variants, nearly all alternative interaction variants assembled more efficiently than the KK / DDD controls. For alternative interaction variants at the E356 / K439 positions, variants showed %MCE purities 3-10% improved compared to the KK / DDD controls except for those with a proline at K439 in the anti-Target 2 half of the heterodimer. %MCE purities for alternative interaction variants at the D399 / K409 positions were even more significantly improved compared to the KK / DDD controls, where purities were improved by 8-22%, with the exception of those with an asparagine at K409 in the antiTarget 2 half.EXAMPLE 4: HETERODIMER THERMOSTABILITY

[0128] To assess whether the mutations affected the thermostability of the assembled heterodimers, Tm and Tagg of the molecules was measured on the UNcle platform. Thermostability was measured for a selected portion of the assembled alternative interaction variants that showed higher titers and assembly efficiency. Each sample (9 pL) was loaded in duplicate in the Uncle cassette and run with a thermal ramp from 25 °C to 95 °C at a ramp speed of 0.3°C / min. Uncle Analysis Software Version 4.0 as used to calculate the Tm of each sample using the first derivative of the barycentric mean (BCM) of the fluorescence intensity. The Tagg of each sample was calculated using the intensity of scattered light at 266 nm.As shown in Figure, 8 most variants showed slightly lower Tms (-1-4°C) compared to the KK / DDD controls, although several alternative interaction variants showed reduced Tms up to 7°C lower than the KK / DDD controls. Tagg measures the onset of protein aggregation, and Tm measures the onset of denaturation, and both Tagg and Tm relate to protein stability. Changes in Tagg was negligible for all variants. While the K370E mutation did not affect correct HC assembly in the heterodimer product (Figure 6a), it reduced the Tm by 3.7°C compared to the KK / DDD heterodimer control in the heterodimer containing Target 1 binder 1 and did not affect the Tm in the heterodimer containing Target 1 binder 2. The 3.7°C decrease in Tm for K370E is a slight destabilization compared to the KK / DDD heterodimer control. Therefore the introduction of an additional negatively charged residue in the assembled heterodimer does slightly destabilize the heterodimer.EXAMPLE 5: HETERODIMER PURITY

[0129] Five variants were tested at larger scale in a heterodimer with different targets (target 3 and target 4). The target 3 KK and target 4 DDD homodimers were cloned into pBMV2.1 vectors as bicistronic constructs with the configuration GAPDH-LC-GAPDH-HC. HCs were constructed in IgGl with SEFL2.2 (R309C / N314G / V321C) mutations in the CH2 domains, along with CPMs in the CH3 domains (E356K / D399K for KK homodimers and K392D / K409D / K439D for DDD homodimers) and either the K370E repacking mutation or alternative interaction mutations at K439 and E356 CPM positions. After sequence confirmation, DNA was prepared using Maxi plasmid purification kits, and the homodimers were expressed in a stable CHO cell line expression system at IL scale after transfection via Lipofectamine LTX (Gibco #15338-100). After selection and scaling of cultures, cultured media was harvested after 10 days of production. The homodimers were purified from the cultured media through a two-column purification process where Protein A capture (Mab Select SuRe) was followed by cation exchange chromatography (CEX)(Capto SP ImpRes). Redox assembly of the KK and DDD homodimers into heterodimers was performed as described previously for Target 1 / Target 2 heterodimers. The purity of the correctly assembled heterodimers were then analyzed by non-reducing MCE.

[0130] As shown in Figure 9, yields of the target 4 DDD homodimers were consistent across the K370E repacking variant and the four alternative interaction variants tested. Yields of the KK homodimers varied, and the K439T alternative interaction variant and the K370E KK repacking variant showed the highest yields. As seen previously for the Target 1 / Target 2 redox-assembled heterodimers, the target 3 / target 4 heterodimers assembled with similarly high efficiencies with 95-96% representing correctly assembled heterodimer products for all variants tested.SEQUENCESSEQ ID NO: 1 - HUMAN GERMLINE IGG1 CH3:GQPREPQVYTLPPSRDELTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGKSEQ ID NO: 2 - HUMAN GERMLINE IGG1 CH3:GQPREPQVYTLPPSREEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGKSEQ ID NO: 3 - K439Q, K409R:GQPREPQVYTLPPSREEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSRLTVDKSRWQQGNVFSCSVMHEALHNHYTQQSLSLSPGKSEQ ID NO: 4 - K439T, K409R:GQPREPQVYTLPPSREEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSRLTVDKSRWQQGNVFSCSVMHEALHNHYTQTSLSLSPGKSEQ ID NO: 5 - K439Q, K409R:GQPREPQVYTLPPSREEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSRLTVDKSRWQQGNVFSCSVMHEALHNHYTQQSLSLSPGKSEQ ID NO: 6 - K439L, K409R:GQPREPQVYTLPPSREEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSRLTVDKSRWQQGNVFSCSVMHEALHNHYTQLSLSLSPGKSEQ ID NO: 7 - K392D, K409D, K439D:GQPREPQVYTLPPSREEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYDTTPPVLDSDGSFFLYSDLTVDKSRWQQGNVFSCSVMHEALHNHYTQDSLSLSPGKSEQ ID NO: 8 - E356T, K409R:GQPREPQVYTLPPSRTEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSRLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGKSEQ ID NO: 9 - E356N, K409R:GQPREPQVYTLPPSRNEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSRLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGKSEQ ID NO: 10 - E356I, K409R:GQPREPQVYTLPPSRIEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSRLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGKSEQ ID NO: 11 - E356K, D399K, K370E:GQPREPQVYTLPPSRKEMTKNQVSLTCLVEGFYPSDIAVEWESNGQPENNYKTTPPVLKSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGKSEQ ID NO: 12 - E356K, D399K:GQPREPQVYTLPPSRKEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLKSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGK

Claims

CLAIMSWhat is claimed:

1. A method for producing a heterodimer, wherein the method comprises: a. expressing a first homodimer in a first cell, and expressing a second homodimer in a second cell; b. purifying the first and second homodimers, wherein the first and second homodimers are combined together either before purifying or after purifying; c. subjecting the combined first and second homodimers to reducing conditions; d. subjecting the combined first and second homodimers to oxidizing conditions; and e. obtaining the heterodimer; and wherein the heterodimer comprises one half of the first homodimer (first homodimer half) and one half of the second homodimer (second homodimer half), and wherein the heterodimer comprises the following mutations according to EU numbering: a. E356K and D399K and one of K370E, K370L, K370M, K370Q, and K370T in the first homodimer half; and K392D, K409D, and K439D in the second homodimer half; b . K409R and one of D399 A, D3991, D399L, D399M, D399N, D399P, D399Q, D399S, D399T, and D399V in the first homodimer half; and one of K409A, K409I, K409L, K409M, K409N, K409P, K409Q, K409S, K409T, and K409V in the second homodimer half; c. K409R and one of E356A, E356I, E356L, E356M, E356N, E356P, E356Q, E356S, E356T, and E356V in the first homodimer half; and K409R and one of K439A, K439I, K439L, K439M, K439N, K439P, K439Q, K439S, K439T, K439V in the second homodimer half; d. D399K in the first homodimer half; and K392D and K409D in the second homodimer half; or e. K439D and D399K in the first homodimer half; and E356K, K392D, and K409D in the second homodimer half.

2. The method of Claim 1, wherein the heterodimer comprises E356K, D399K, and K370E mutations in the first homodimer half; and K392D, K409D, and K439D mutations in the second homodimer half.

3. The method of Claim 1, wherein the heterodimer comprises E356N and K409R mutations in the first homodimer half; and K439Q and K409R mutations in the second homodimer half.

4. The method of Claim 1, wherein the heterodimer comprises E356T and K409R mutations in the first homodimer half; and K439Q and K409R mutations in the second homodimer half.

5. The method of Claim 1, wherein the heterodimer comprises E356T and K409R mutations in the first homodimer half; and K439T and K409R mutations in the second homodimer half.

6. The method of any one of Claims 1-5, wherein the first homodimer and / or the second homodimer are produced at a higher yield compared to a homodimer comprising only E356K and D399K mutations or a homodimer comprising only K392D, K409D, and K439D mutations.

7. The method of any one of Claims 1-6, wherein the reducing conditions comprise mixing homodimers with about 5 mM GSH, resulting in a reduced mixture.

8. The method of Claim 7, wherein the reduced mixture is incubated at about 36°C for about five hours.

9. The method of Claim 8, wherein after incubating for about five hours, the reduced mixture is buffer exchanged into IX PBS, pH 7.4 and incubated for 12-16 hours at about 36°C.

10. The method of any one of Claims 1-6, wherein the reducing conditions comprise mixing homodimers with about 25 mM 2-MEA, resulting in a reduced mixture.

11. The method of Claim 10, wherein the reduced mixture is incubated at about 22°C for at least about 5 hours.

12. The method of Claim 11, wherein after incubating, the reduced mixture is buffer exchanged into Tris pH 7.4 to a final concentration of about 100 mM, and incubated at about 22°C for 12-16 hours.

13. A heterodimer produced according to any one of Claims 1-12.

14. A heterodimer that comprises two homodimer halves, the heterodimer comprising the following mutations according to EU numbering: a. E356K and D399K and one of K370E, K370L, K370M, K370Q, and K370T in one homodimer half of the heterodimer, and K392D, K409D, and K439D in the other homodimer half of the heterodimer;b . K409R and one of D399 A, D3991, D399L, D399M, D399N, D399P, D399Q, D399S, D399T, and D399V in one homodimer half of the heterodimer; and one of K409A, K409I, K409L, K409M, K409N, K409P, K409Q, K409S, K409T, and K409V in the other homodimer half of the heterodimer; c. K409R and one of E356A, E356I, E356L, E356M, E356N, E356P, E356Q, E356S, E356T, and E356V in one homodimer half of the heterodimer; and K409R and one of K439A, K439I, K439L, K439M, K439N, K439P, K439Q, K439S, K439T, and K439V in the other homodimer half of the heterodimer; d. D399K in one homodimer half of the heterodimer; and K392D and K409D in the other homodimer half of the heterodimer; or e. K439D and D399K in one homodimer half of the heterodimer; and E356K,K392D, and K409D in the other homodimer half of the heterodimer.

15. A pharmaceutical composition comprising the heterodimer of Claim 13 or 14.

Images