Chemoselective reduction of cysteine-engineered antibodies for preparation of antibody drug conjugates

Abstract

Methods for making antibody-drug conjugates (ADCs) from antibodies that comprise two or more engineered cysteines per antibody conjugated to one or more payloads using phosphine-based reducing agents that are capable of selectively reducing engineered cysteine pairs on the antibody backbone are described.

Description

CHEM0SELECT1VE REDUCTION OF CYSTEINE-ENGINEERED ANTIBODIES FORPREPARATION OF ANTIBODY DRUG CONJUGATESCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Patent Application Serial No. 63 / 661,205 filed June 18, 2024, the entire contents of which are incorporated by reference herein.FIELD

[0002] This disclosure relates generally to methods for making antibody-drug conjugates (ADCs) from antibodies that comprise two or more engineered cysteines per antibody conjugated to one or more payloads using phosphine-based reducing agents capable of selectively reducing engineered cysteines on the antibody backbone.BACKGROUND

[0003] ADCs have revolutionized the field of cancer therapy and had meaningful impact on patient lives (Dumontet et al., Nat. Rev. Drug Discov. 22: 641-661 (2023); Zinn et al., Nat. Cancer 4: 165-180 (2023)). Currently, there are 14 ADCs approved for the treatment of solid tumors and heme malignancies and over 120 ADCs are currently being evaluated in clinical trials across various stages (Maecker et al., MAbs 15: 2229101 (2023)). Despite their clinical success, ongoing efforts are focused on advancing conjugation technology, linker chemistry, antibody engineering and payload types to further broaden the therapeutic window (Colombo and Rich, Cancer Cell 40: 1255-1263 (2022)). As a part of these efforts, site-specific ADCs have emerged as a next generation of conjugates, offering significant improvement over the limitations associated with current ADCs (Chudasama et al., Nat Chem 8: 114-119 (2016); Hoffmann et al., Oncolmmunology 7: el395127 (2018); Junutula et al., Clin.l Can. Res. 16: 4769-4778 (2010); Lyons et al., Prot. Engineer., Des. Selec. 3: 703-708 (1990); Ohri et al., Bioconj. Chem 29: 473- 485 (2018); Su et al., Bioconjug. Chem. 29: 1155-1167 (2018)).

[0004] Conventional ADCs are typically produced by conjugating the linker drug to the antibody through the side chains of either surface-exposed lysines or free cysteines generated through reduction of interchain disulfide bonds. Because antibodies contain many lysine residues and cysteine disulfide bonds, conventional conjugation typically produces heterogeneous mixtures that present challenges with respect to analytical characterization and manufacturing. Furthermore, the individual constituents of these mixtures exhibit different physicochemicalproperties and pharmacology with respect to their pharmacokinetic, efficacy, and safety profiles, hindering a rational approach to optimizing this modality.

[0005] The site-specific introduction by genetic engineering of a cysteine residue into a suitable position in the antibody allows control of the site of conjugation and the obtained site-specific conjugates (or site-specific ADCs) are more homogeneous than the conjugates obtained via wildtype conjugation, i.e., conjugation via interchain disulfide cysteines. Such wild-type conjugation leads to a heterogeneous mixture of conjugates, which is especially disadvantageous in the case of ADCs. Some individual constituents of a wild-type conjugated ADC mixture can have poor in vivo performance. The in vivo performance of ADCs in terms of efficacy, safety, and stability may be improved if the linker drugs of the ADCs are site-specifically conjugated via engineered cysteines according to Shen et al., Nature Biotechnology, 2012, 30 (2), 184-189. Methods for site-specific drug conjugation to antibodies are comprehensively reviewed by Behrens et al., MAbs, 2014. 6(1), 1-8. The first site-specific conjugation approach was developed at Genentech by introducing an engineered cysteine residue using site-directed mutagenesis at positions showing high thiol reactivity as elaborated in W02006034488. This common practice in protein modification is more complicated in an antibody because of the various native cysteine residues already present. Introducing the extra cysteine residue in an unsuitable position could result inter alia in improper formation of interchain disulfide bonds and therefore improper folding of the antibody.

[0006] Site-specific ADCs have demonstrated notable improvements in biophysical properties, reduced molecular heterogeneity, and improved toxicity profile compared to ADC prepared via stochastic conjugations (Lyons et al., op. cit.; Junutula et al., Nat. Biotechnol. 26: 925-932 (2008); Strop et al., Chem. & Biol. 20: 161-167 (2013); Strop et al., Mol. Can. Thera. 15: 2698- 2708 (2016)). The ability to identify unique conjugation sites on the antibody that offers these advantages has grown significantly (Jeffrey et al., Bioconjug. Chem. 24: 1256-1263 (2013); Li et al., Mol. Cane. Thera. 17: 554-564 (2018); Tumey et al., The AAPS Journal 19: 1123-1135 (2017)). However, preparation of high-quality site-specific ADCs is not simple and encounters significant technical challenges. Site-specific ADC processing, using engineered cysteines, commonly involves multiple steps, including harsh reduction conditions for de-capping of engineered cysteine and an imperfect reoxidation step to reform inter-chain disulfide bonds. This multistep conjugation method adds complexity and introduces undesired side-products (Fig. 1A) such as scrambled disulfides, overoxidation and half-antibody fragments (Coumans et al., Bioconjug. Chem. 31, 2136-2146 (2020); Procopio-Melino et al., Scien. Reports 12: 7262 (2022)).

[0007] To address these challenges, selective reduction of engineered cysteines has been explored as a practical solution by many research groups. For example, U.S. Patent Nos. 11,667,696 and 10,464,997 demonstrated selective reduction of engineered cysteines of an antibody at heavy chain positions 326. 337, and 239 (Eu numbering) by using low concentrations of cysteine without a reoxidation step. However, this process does not work against all engineered cysteine sites and often requires the removal of excess reductant before adding the reactive linker-payload. WO2015177360 and U.S. Patent Nos. 10,814,009 and 11,696,958 demonstrated selective reduction of engineered cysteines of an antibody at one or more of heavy chain positions 40. 41, 42, or 89 (Kabat numbering); heavy chain positions 152, 153. 155, or 171 (Eu numbering); light chain positions 40 or 41 (Kabat numbering); or light chain positions 165 or 168 (Eu numbering) by using a selection of phosphine-based reductants to produce reduced cysteines at the one or positions of engineered cysteines.

[0008] In another example, U.S. Patent Application Publication No. 2021017222 discloses use of various phosphine-based reductants for selective reduction and conjugation of an IgGl antibody engineered to lack the native cysteine residues at position 220 of the heavy chains or 214 of the light chains (Eu numbering). The cysteines at these positions normally form a disulfide linkage between the heavy chain and light chain to stabilize the heavy chain: light chain pairing. While replacing the cysteine at one of the positions permits the use of the cysteine at the other position for conjugation to a payload comprising a cysteine reactive group, the structural integrity of the antibody is not the same as that of antibodies that maintain the disulfide bond.

[0009] Using bulky and hydrophobic triphenyl phosphine reducing agents for selective reduction of engineered cysteines has been investigated by few researchers. Zhong et al., demonstrated selective reduction of thio-2-nitrobenzoic acid (TNB) (Procopio-Melino et al., op. cit.) capped engineered cysteine at positions 183 and 290 of trastuzumab (Eu numbering) with Triphenylphosphine-3,3',3''-trisulfonic acid trisodium salt. The effectiveness of this method for engineered cysteine sites other than the ones reported is unknown. Additionally, achieving TNB capping for every new antibody target would require metabolic engineering of cells, potentially limiting its broad applicability.

[0010] Coumans et al., op. cit., and Liao et al. (Bioconjug. Chem. 34: 2293-2301 (2023)) employed a screening approach using a small library of triphenylphosphines to identify reagents that can selectively reduce unpaired cysteines in either the Fab pocket or the antibody hinge region, while preserving the integrity of interchain bonds in the hinge region. These methods have proven capable of generating homogeneous and site-specific ADCs through a mild, one-pot process.

[0011] Although these reports point to advancements in preparation of site-specific ADCs, there are improvements needed for mainstream adoption of this chemistry. This is mainly due to lack of mechanistic understanding behind selectivity' of phosphine-based reductants for reducing specific engineered cysteine sites.SUMMARY

[0012] The present disclosure provides a broadly applicable strategy for identification and design of reductants capable of selectively reducing capped engineered cysteines at any site on the antibody for making antibody-drug conjugates (ADCs). Computational modeling and reduction rate kinetics of phosphine-based reductants were investigated to gain extensive insights into selective reduction mechanism. The methodology disclosed herein offers a comprehensive framework for the chemoselective reduction of a wide range of engineered sites, facilitating accelerated early and late-stage ADC drug discovery.

[0013] Using methodologies herein, phosphine-based reductants capable of selectively reducing capped engineered cysteines on the antibody backbone over the interchain and intrachain disulfide bonds of the antibody have been identified and methods have been developed using these reductants to efficiently produce compositions of ADCs in which greater than 90% of the antibodies comprising the ADCs are intact and with drug-to-antibody ratios (DARs) consistent with substantially complete conjugation of the available engineered cysteine residues. The methodologies disclosed herein further enabled the identification of reductant pairs and reducing conditions that permit selective reduction of capped engineered cysteine residues at one position in the antibody over another position in the antibody resulting in the ability to produce ADCs conjugated to two separate payloads.

[0014] The present invention provides a method for selectively reducing a capped engineered cysteine (Cys) residue of a Cys-engineered antibody over reducing interchain disulfide bonds to produce a reduced, uncapped Cys-engineered antibody, the method comprising contacting the Cys-engineered antibody with a reductant selected from the group consisting of: 4- (diphenylphosphaneyl)benzenesulfonic acid (2), 4,4'-(phenylphosphanediyl)dibenzenesulfonic acid (3), 3-(diphenylphosphaneyl)benzenesulfonic acid (4), 4-(diphenylphosphaneyl)benzoic acid (5), 4-(diphenylphosphaneyl)-N,N-dimethylaniline (6), (R)-2-((diphenylphosphaneyl)methyl)pyrrolidine (7), 2-(diphenylphosphaneyl)benzenesulfonic acid (8), triphenylphosphine (9), 2-(diphenylphosphaneyl)benzaldehyde (10), tris(4- methoxyphenyl)phosphane (11), 2-(di-p-tolylphosphaneyl)benzaldehyde (12), 2- (dicyclohexylphosphaneyl)benzenesulfonic acid (13), 4-(bis(4-fluorophenyl)phosphaneyl)benzenesulfonic acid (24). tris(4-fluorophenyl)phosphane (28). 4- (diphenylphosphaneyl)-N,N,N-trimethylbenzenaminium (29), 4-((4- fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30), and salts thereof, for a time sufficient to reduce the capped engineered cysteine of the Cys -engineered antibody to produce the reduced, uncapped Cys-engineered antibody.

[0015] In further embodiments of the method, the Cys-engineered antibody comprises one or more engineered cysteine residues, wherein each engineered cysteine residue is located at a position within the Cys-engineered antibody selected from the group consisting of heavy chain positions 152, 153. 171, 172, 173, 347, 362. 375, 380. 382, 393, 430, and 440; wherein the position number is according to EU numbering; and / or the group consisting of heavy chain positions 40, 41, and 89; wherein the position number is according to Kabat numbering.

[0016] In further embodiments of the method, the Cys-engineered antibody comprises one or more engineered cysteine residues, wherein each engineered cysteine residue is located at a position within the Cys-engineered antibody selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering; and / or the group consisting of light chain positions 40 and 41; wherein the position number is according to Kabat numbering.

[0017] In further embodiments of the method, the Cys-engineered antibody comprises (a) a first engineered cysteine residue located at a position selected from the group consisting of heavy chain positions 152, 153, 171, 172, 173, 347, 362, 375, 380, 382, 393, 430, and 440; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue located at a position selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering.

[0018] In further embodiments of the method, the Cys-engineered antibody comprises (a) a first engineered cysteine residue at heavy’ chain position 375; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue at light chain position 168; wherein the position number is according to EU numbering.

[0019] The present invention further provides a method for producing a composition of antibody conjugates, the method comprising: (a) providing a composition of Cys-engineered antibodies, wherein each antibody comprises (i) two heavy chains and two light chains, (ii) one or more capped engineered cysteine residues, and (iii) interchain disulfide bonds between the hinge region of the two heavy chains and between the hinge region of the heavy chain and the light chain; (b) mixing the composition with an excess of reductant selected from the group consisting of: 4-(diphenylphosphaneyl)benzenesulfonic acid (2), 4,4'-(phenylphosphanediyl)dibenzenesulfonic acid (3). 3-(diphenylphosphaneyl)benzenesulfonic acid (4), 4-(diphenylphosphaneyl)benzoic acid (5), 4-(diphenylphosphaneyl)-N,N-dimethylaniline (6), (R)-2-((diphenylphosphaneyl)methyl)pyrrolidine (7), 2-(diphenylphosphaneyl)benzenesulfonic acid (8), triphenylphosphine (9), 2-(diphenylphosphaneyl)benzaldehyde (10), tris(4- methoxyphenyl)phosphane (11). 2-(di-p-tolylphosphaneyl)benzaldehyde (12), 2- (dicyclohexylphosphaneyl)benzenesulfonic acid (13), 4-(bis(4- fluorophenyl)phosphaneyl)benzenesulfbnic acid (24), tris(4-fluorophenyl)phosphane (28), 4- (diphenylphosphaneyl)-N,N,N-trimethylbenzenaminium (29), 4-((4- fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30), and salts thereof, to provide a reaction mixture; (c) incubating the reaction mixture for a time sufficient to selectively reduce the capped engineered cysteines of the Cys-engineered antibodies over reducing the interchain disulfide bonds of the Cys-engineered antibodies to produce reduced, uncapped Cys-engineered antibodies; (d) separating the reduced, uncapped Cys-engineered antibodies from the reductant to provide a composition of separated, reduced, uncapped Cys-engineered antibodies: and (e) incubating the composition of separated, reduced, uncapped Cys-engineered antibodies with an excess of payload comprising a thiol reactive group for a time sufficient to produce the composition of antibody conjugates.

[0020] In further embodiments of the method, the Cys-engineered antibody comprises one or more engineered cysteine residues, wherein each engineered cysteine residue is located at a position within the Cys-engineered antibody selected from the group consisting of heavy chain positions 152, 153, 171, 172, 173, 347, 362, 375, 380. 382, 393, 430, and 440; wherein the position number is according to EU numbering; and / or the group consisting of heavy chain positions 40. 41. and 89; wherein the position number is according to Kabat numbering.

[0021] In further embodiments of the method, the Cys-engineered antibody comprises one or more engineered cysteine residues, wherein each engineered cysteine residue is located at a position within the Cys-engineered antibody selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering; and / or the group consisting of light chain positions 40 and 41; wherein the position number is according to Kabat numbering.

[0022] In further embodiments of the method, the Cys-engineered antibody comprises (a) a first engineered cysteine residue located at a position selected from the group consisting of heavy chain positions 152, 153, 171, 172, 173, 347, 362, 375, 380, 382, 393, 430, and 440; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residuelocated at a position selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering.

[0023] In further embodiments of the method, the Cys-engineered antibody comprises (a) a first engineered cysteine residue at heavy chain position 375; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue at light chain position 168; wherein the position number is according to EU numbering.

[0024] In further embodiments of the method, the excess of reductant comprises 6 equivalences of reductant per engineered cysteine.

[0025] In further embodiments of the method, the excess of pay load comprises 4 equivalences of payload per engineered cysteine.

[0026] In further embodiments of the method, greater than 90% of the Cys-engineered antibodies in the composition of antibody conjugates are intact following reduction and conjugation.

[0027] In further embodiments of the method, the composition of antibody conjugates has a drug-to-antibody ratio (DAR) between 1.7 and 2.0.

[0028] The present invention further provides a method for producing a composition of antibody conjugates conjugated to a first and a second payload, the method comprising: (a) providing a mixture comprising Cys-engineered antibodies having a heavy chain and a light chain, a first capped engineered cysteine residue at position 168 of the light chain, a second capped engineered cysteine residue at position 375 of the heavy chain, and interchain disulfide bonds, wherein the position numbers are according to EU numbering; (b) adding to the mixture an excess of first reductant 2-(diphenylphosphaneyl)benzenesulfonic acid (8) or salt thereof to produce a first reducing mixture, and incubating the first reducing mixture for a time sufficient to selectively reduce the first capped engineered cysteines at position 168 of the light chains without reducing the interchain disulfide bonds of the Cys-engineered antibodies; (c) separating the first reduced, uncapped Cys-engineered antibodies from the first reductant to provide a first composition of Cys-engineered antibodies; (d) adding an excess of a first payload comprising a thiol reactive group to the first composition to produce a first conjugation mixture and incubating the first conjugation mixture for a time sufficient to produce Cys-engineered antibodies conjugated to the first payload; (f) separating the Cys-engineered antibodies conjugated to the first payload from the first conjugation mixture to provide a composition of Cys-engineered antibodies conjugated to the first payload; (g) adding to the composition of Cys-engineered antibodies conjugated to the first payload an excess of a second reductant 4- (diphenylphosphaneyl)benzenesulfonic acid (2) or salt thereof to produce a second reducingmixture, and incubating the second reducing mixture for a time sufficient to selectively reduce the second capped engineered cysteines at position 375 of the heavy chains without reducing the interchain disulfide bonds of the Cys-engineered antibodies to produce second reduced, uncapped Cys -engineered antibodies; (h) separating the second reduced, uncapped Cys-engineered antibodies from the second reductant to provide a second composition of Cys-engineered antibodies; (i) adding an excess of a second payload comprising a thiol reactive group to the second composition to produce a second conjugation mixture and incubating the second conjugation mixture for a time sufficient to produce Cys-engineered antibodies conjugated to the second payload; and (h) separating the Cys-engineered antibodies conjugated to the second payload from the second conjugation mixture to provide a composition of Cys-engineered antibodies conjugated to the first payload and the second payload.

[0029] In a further embodiment of the method, the excess of the first reductant comprises about 40 equivalences of reductant per capped engineered cysteine.

[0030] In a further embodiment of the method, the excess of the second reductant comprises about 6 equivalences of reductant per capped engineered cysteine.

[0031] In a further embodiment of the method, the excess of pay load comprises 4 equivalences of payload per reduced engineered cysteine.

[0032] The present invention further provides a dual-payload antibody conjugate comprising an antibody having a light chain comprising a cysteine residue at position 168 conjugated to a first payload and a heavy chain comprising a cysteine residue at position 375 conjugated to a second payload, wherein the position numbering is according to Eu numbering and wherein the first payload and the second payload are not the same.

[0033] The present invention further provides an antibody conjugate comprising an antibody having a light chain and a heavy chain, wherein the light chain comprises an engineered cysteine at position 168 conjugated to a first payload and the heavy chain comprises an engineered cysteine residue at position 375 conjugated to a second payload, and wherein the position numbering is according to Eu numbering and wherein the first payload and the second payload are not the same.

[0034] The present invention further provides phosphine-based reductants that selectively reduce engineered cysteines of an antibody over reducing interchain disulfide bonds, wherein greater than 90% of the antibodies remain intact following reduction. Exemplary phosphinebased reductants may be selected from the group consisting of: 4-(bis(4- fluorophenyl)phosphaneyl)benzenesulfonic acid (24), 4-(diphenylphosphaneyl)-N,N,N-trimethylbenzenaminium (29). 4-((4-fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30), and salts thereof.

[0035] The present invention further provides a reductant selected from the group consisting of: 4-(dicyclohexylphosphaneyl)benzoic acid (22), 4-(dicyclohexylphosphaneyl)benzenesulfonic acid (23), 4-(bis(4-methoxyphenyl)phosphaneyl)benzenesulfonic acid (25). 4- (cyclohexyl(phenyl)phosphaneyl)benzoic acid (26), 4- (cyclohexyl(phenyl)phosphaneyl)benzenesulfonic acid (27), and salts thereof.

[0036] The summary of the technology' described above is non-limiting and other features and advantages of the technology’ yvill be apparent from the following detailed description, and from the claims.BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Fig. 1A shows the challenges of engineered cysteine conjugation in ADCs using global full reduction and oxidation.

[0038] Fig. IB shows a general strategy to prepare ADCs based on screening, accessibility, and reactivity of phosphine-based reagents through computational and experimental approaches.

[0039] Fig. 2A - Fig. 2E. Screening methodology for commercially relevant triphenylphosphine.

[0040] Fig. 2A shows the structures of triphenylphosphines screened in the study and their classification according to experimental and computational selectivity towards S375C and S168C.

[0041] Fig. 2B shows S375C and S168C engineered IgGl antibody (based on cry stal the structure of human antibody IgGl bl2; Protein Data Bank (PDB) structure ID: 1HZH).

[0042] Fig. 2C shows a reaction scheme demonstrating stepwise reduction and conjugation at S375C or S168C engineered antibody,

[0043] Fig. 2D and Fig. 2E show LCMS traces of IdeS digestion and full reduction using TCEP revealed complete conjugation with reductant 19 at Fc or LC site yvithout modifying interchain disulfides.

[0044] Fig. 3A - Fig. 3F. Computational characterization of phosphine selective reduction and reduction rate kinetics of the phosphine reductants.

[0045] Fig. 3A shows pKa prediction of engineered and interchain disulfides suggesting that the cysteines of interest have similar reactivity'. Averages and standard deviations from structural ensemble are reported.

[0046] Fig. 3B. shows solvent accessible surface area of sulfur atoms in disulfides of interest from 100 ns MD simulation. Averages and standard deviations from structural ensemble are reported.

[0047] Fig. 3C shows the combined structure-based approach to represent selectivity as relative reagent accessibility. The diagrams illustrate the details of the computational workflow. The correlation between computed accessibility difference and observed selective / non-selective behavior of reductants respectively towards S375C and S168C is show n at the bottom.

[0048] Fig. 3D shows reductants (22-30), which w ere designed following accessibility model and structure reactivity relationship (SAR) studies described herein.

[0049] Fig. 3E shows the mechanism of disulfide reduction as illustrated by formation of thiophosphonium ions as a rate limiting step followed by rapid hydrolysis of the complex to release the second thiol fragment and the phosphine oxide.

[0050] Fig. 3F shows P NMR arrays, which reflect phosphine exhaustion rate for incomplete decapping using reductant 1. site-specific decapping by reductant 5, and non-selective decapping by reductant 16.

[0051] Fig. 4A - Fig. 4D show' dual-payload ADC synthesis using orthogonal reactivity of phosphine reagents.

[0052] Fig. 4A shows that an excess amount of reductant 8 led to S168C reduction and conjugation with 19 while reduction of S375C by reductant 2 that was conjugated to 21 in sequential manner under mild reaction conditions.

[0053] Fig. 4B shows the structure of linker-payloadss 19 (comprising drug MMAE and 21 (comprising the drug monomethylauristatin F (MMAF)) that were used for conjugation.

[0054] Fig. 4C LCMS of IdeS digestion and full reduction using TCEP revealed selectivity at two different sites of conjugation.

[0055] Fig. 4D shows that intact dual-drug ADC w as 98% monomeric without causing aggregation.

[0056] Fig. 5 is a cartoon showing the relative positions of the intrachain disulfide bonds and interchain disulfide bonds of an IgGl antibody and an IgG4 antibody.

[0057] Fig. 6A LCMS show s that reductant 10 stalled at thiophosphonium intermediate state.

[0058] Fig. 6B LCMS shows that reductant 12 stalled at thiophosphonium intermediate state.

[0059] Fig. 7 shows a31P NMR array used to determine reduction kinetics for reductant 1.

[0060] Fig. 8 shows a31P NMR array used to determine reduction kinetics for reductant 2.

[0061] Fig. 9 shows a31P NMR array used to determine reduction kinetics for reductant 3.

[0062] Fig. 10 shows a31P NMR array used to determine reduction kinetics for reductant 4.

[0063] Fig. 11 shows a31P NMR array used to determine reduction kinetics for reductant 5.

[0064] Fig. 12 shows a31P NMR array used to determine reduction kinetics for reductant 6.

[0065] Fig. 13 shows a31P NMR array used to determine reduction kinetics for reductant 8.

[0066] Fig. 14 shows a31P NMR array used to determine reduction kinetics for reductant 9.

[0067] Fig. 15 shows a31P NMR array used to determine reduction kinetics for reductant 11.

[0068] Fig. 16 shows a31P NMR array used to determine reduction kinetics for reductant 15.

[0069] Fig. 17 shows a31P NMR array used to determine reduction kinetics for reductant 24.

[0070] Fig. 18 shows a31P NMR array used to determine reduction kinetics for reductant 29.

[0071] Fig. 19 shows a31P NMR array used to determine reduction kinetics for reductant 30.

[0072] Fig. 20A - Fig. 20C shows the development of the reductant accessibility evaluation workflow.

[0073] Fig. 20A shows comparison of computational accessibility' descriptors after small molecule docking and after constrained evaluation for S375C, shows that constrained relaxation resulted in improved differentiation of selective / non-selective phosphines for H-H site.

[0074] Fig. 20B shows comparison of computational accessibility' descriptors after small molecule docking and after constrained evaluation for S168C, shows that constrained relaxation resulted in improved differentiation of selective / non-selective phosphines for H-H site.

[0075] Fig. 20C shows for example, the correlation between different accessibility descriptors respectively after docking and constrained relaxation indicates that a shared physical property was represented by both descriptors.

[0076] Figs. 21A - Fig. 21B shows selective decapping and conjugation at position S375C and S168C of an IgGl antibody.

[0077] Fig. 21A shows selective decapping and conjugation at position S375C of an IgGl antibody using 20 equivalences of reductant 1 followed by conjugation w ith compound 19. The data. Note: Excess reductant w as removed prior to adding 19.

[0078] Fig. 21B shows selective decapping and conjugation at position S375C of an IgGl antibody using 20 equivalences of reductant 9 followed by conjugation w ith compound 19. The data. Note: Excess reductant w as removed prior to adding compound 19.DETAILED DESCRIPTION

[0079] Site-specific ADCs have demonstrated notable improvements in stability and toxicity compared to analogous ADC prepared via conventional routes. However, preparation of sitespecific ADCs employing engineered cysteines is complex and results in undesired side products. Recent approaches utilizing tripheny lphosphine derivatives for selective reduction of engineeredcysteines have limited applicability and their use suffer from a lack of comprehensive understanding of the chemoselective reduction process. The present disclosure provides a broadly applicable strategy for identification and design of reductants capable of selectively reducing engineered cysteines at any site on the antibody. Computational modeling and reduction rate kinetics of phosphine-based reductants were investigated to gain extensive insights into the selective reduction mechanism. The methodology disclosed herein offers a comprehensive framework for the chemoselective reduction of a wide range of engineered sites, facilitating accelerated early and late-stage ADC drug discover}'.Definitions

[0080] Listed below are definitions of various terms used herein. These definitions apply to the terms as they are used throughout this specification and claims, unless otherwise limited in specific instances, either individually or as part of a larger group.

[0081] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. Generally, the nomenclature used herein and the laboratory' procedures in cell culture, molecular genetics, organic chemistry, and peptide chemistry are those well-known and commonly employed in the art.

[0082] As used herein, the articles “a” and "an" refer to one or to more than one (i. e.. to at least one) of the grammatical object of the article. By way of example, “an element” means one element or more than one element. Furthermore, use of the term “including” as well as other forms, such as “include,” “includes,” and “included,” is not limiting.

[0083] As used herein, the term “about” in quantitative terms refers to plus or minus 10% of the value it modifies (rounded up to the nearest whole number if the value is not sub-dividable, such as a number of molecules or nucleotides).

[0084] All ranges disclosed herein are inclusive of the recited endpoint and independently combinable (for example, the range of “from 50 mg to 500 mg” is inclusive of the endpoints, 50 mg and 500 mg, and all the intermediate values). The endpoints of the ranges and any values disclosed herein are not limited to the precise range or value; they are sufficiently imprecise to include values approximating these ranges and / or values.

[0085] As used herein, the term “comprising” may include the embodiments “consisting of’ and “consisting essentially of.” The terms “comprise(s),” “include(s),” “having,” “has,” “may,” “contain(s),” and variants thereof, as used herein, are intended to be open-ended transitional phrases, terms, or words that require the presence of the named ingredients / steps and permit the presence of other ingredients / steps. However, such description should be construed as alsodescribing compositions or processes as “consisting of’ and “consisting essentially of’ the enumerated components, which allows the presence of only the named components or compounds, along with any acceptable carriers or fluids, and excludes other components or compounds.

[0086] As used herein, the term "antibody" or “immunoglobulin” as used herein refers to a glycoprotein comprising at least two heavy chains (HCs) and two light chains (LCs) interconnected by disulfide bonds. A typical IgGl or IgG4 antibody contains two identical LCs and two identical HCs and contains a total of 16 disulfide bonds, including four inter-chain disulfide bonds in the hinge region of the HC and 12 intra-chain disulfide bonds associated with 12 individual domains of the antibody. As shown in Fig. 5, there are two inter-chain disulfide bonds that connect the cysteine residue of one hinge region to the other hinge region and there is one disulfide bond that connects a cy steine residue in the constant domain of one light chain to the cysteine residue in one hinge region and one disulfide bond that connects a cysteine residue in the constant domain of the other light chain to the cysteine residue in the other hinge region. Each HC is comprised of a heavy chain variable region or domain (VH) and a heavy chain constant region or domain. Each light chain is comprised of an LC variable region or domain (VL) and a LC constant domain. In certain naturally occurring IgG, IgD, and IgA antibodies, the heavy chain constant region is comprised of three domains, CHI. CH2 and CH3. In general, the basic antibody structural unit for antibodies is a Y-shaped tetramer comprising two HC / LC pairs (2H). Each tetramer includes two identical pairs of polypeptide chains, each pair having one LC (about 25 kDa) and HC chain (about 50-70 kDa) (H+L). Each HC:LC pair comprises one VH: one VL pair. The one VH:one VL pair may be referred to by the term “Fab”. Thus, each antibody tetramer comprises two Fabs, one per each arm of the Y-shaped antibody.

[0087] The human VH includes seven family members: VH1, VH2, VH3, VH4, VH5, VH6, and VH7; and the human VL includes 16 family members: VKI, VK2, VK3, VK4, VK5, VK6, Vkl, VZ2. V / .3. V / .4. VZ5. V / .6. VX7, V / .8. V / .9. and VZ 10. Each of these family members can be further divided into particular subtypes. The VH and VL can be further subdivided into regions of hypervariability, termed complementarity7determining region (CDR) areas, interspersed with regions that are more conserved, termed framework regions (FR). Each VH and VL is composed of three CDR regions and four FR regions, arranged from amino-terminus to carboxy -terminus in the following order: FR1, CDR 1, FR2, CDR 2, FR3, CDR 3. FR4. Numbering of the amino acids in a Vjq and VL may be determined using the Kabat numbering scheme. See Beranger et al., Ed. Ginetoux, Correspondence between the IMGT unique numbering for C-DOMAIN, the IMGT exon numbering, the Eu and Kabat numberings: Human IGHG, created: 17 / 05 / 2001,Version: 08 / 06 / 2016, which is accessible at www.imgt.org / lMGTScientificChart / Numbering / Hu_IGHGnber.html).

[0088] The constant regions of the antibodies may mediate the binding of the immunoglobulin to host tissues or factors, including various cells of the immune system (e.g., effector cells) and the first component (Clq) of the classical complement system. Typically, the numbering of the amino acids in the heavy chain constant domain begins with number 118, which is in accordance with the Eu numbering scheme. The Eu numbering scheme is based upon the amino acid sequence of human IgGl (Eu). w hich has a constant domain that begins at amino acid position 118 of the amino acid sequence of the IgGl described in Edelman et al., Proc. Natl. Acad. Sci. USA. 63: 78-85 (1969), and is shown for the IgGl, IgG2, IgG3, and IgG4 constant domains in Beranger et al., op. cit.

[0089] The variable regions of the heavy and light chains contain a binding domain comprising the CDRs that interacts with an antigen. A number of methods are available in the art for defining CDR sequences of antibody variable domains (see Dondelinger et al., Frontiers in Immunol. 9: Article 2278 (2018)). The common variable region numbering schemes include the following.

[0090] Kabat numbering scheme is based on sequence variability and is the most commonly used (See Kabat et al. Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, Md. (1991) (defining the CDR regions of an antibody by sequence).

[0091] Chothia numbering scheme is based on the location of the structural loop region (See Chothia & Lesk, J. Mol. Biol. 196: 901-917 (1987); Al-Lazikani et al., J. Mol. Biol. 273: 927- 948 (1997)).

[0092] AbM numbering scheme is a compromise between the two used by Oxford Molecular's AbM antibody modelling software (see Kara et al., ILAR Journal 37: 132-141 (1995).

[0093] Contact numbering scheme is based on an analysis of the available complex cry stal structures (See www.bioinf.org.uk: Prof. Andrew C.R. Martin's Group; Abhinandan & Martin. Mol. Immunol. 45:3832-3839 (2008)).

[0094] IMGT (ImMunoGeneTics) numbering scheme is a standardized numbering system for all the protein sequences of the immunoglobulin superfamily, including variable domains from antibody light and heavy chains as w ell as T cell receptor chains from different species and counts residues continuously from 1 to 128 based on the germ-line V sequence alignment (see Giudicelli et al., Nucleic Acids Res. 25:206-1 1 (1997); Lefranc, Immunol Today 18:509(1997); Lefranc et al., Dev Comp Immunol. 27:55-77 (2003)).

[0095] The entire amino acid sequence of the VH and VL are commonly numbered according to Kabat while the three CDRs within the variable region may be defined according to any one of the aforementioned numbering schemes. In particular embodiments, the numbering of the amino acid positions in the VH and VL may be sequential beginning with amino acid position 1 and continuing sequentially to the end of the sequence or according to Kabat. Unless specified otherwise, the amino acid positions in the VH and VL herein are defined according to sequential numbering.

[0096] The numbering of the amino acid positions in the heavy chain constant domain may be sequential beginning with amino acid position 1 and continuing sequentially to the end of the sequence or according to Eu numbering. The IgGl heavy chain constant domain amino acid sequence has 330 amino acids, which may be sequentially numbered 1 to 330. The corresponding sequence numbered according to Eu begins with position number 118 and ends with position number 447. Unless specified otherwise, the amino acid positions in the heavy and light chains herein are defined according to sequential numbering.

[0097] The numbering of the amino acid positions in the light chain constant domain may be sequential beginning with amino acid position 1 and continuing sequentially to the end of the sequence or according to Eu numbering. The IgGl light chain constant domain amino acid sequence has 107 amino acids, which may be sequentially numbered 1 to 107. The corresponding sequence numbered according to Eu begins with position number 108 and ends with position number 214. Unless specified otherwise, the amino acid positions in the heavy and light chains herein are defined according to sequential numbering.

[0098] The term antibody as used in the context of the present invention refers to a Cys- engineered antibody unless specifically otherwise indicated.

[0099] As used herein, the terms “Cys-engineered antibody,” '‘Cys-engineered construct”, and “site-specific antibody” refer to antibodies wherein at least one amino acid in either the heavy and / or light chain is replaced with a cysteine residue to provide at least one unpaired cysteine (the “engineered cysteine residue”). Similarly, a “Cys-engineered conjugate”, “site-specific conjugate”, or “site-specific ADC” refers to an antibody-drug conjugate comprising such engineered antibody and at least one payload conjugated to the unpaired cysteine(s). In certain embodiments, the unpaired cysteine residue will be at position 168 of the light chain (Eu numbering), position 375 of the CH3 region of the heavy constant domain (Eu numbering), or both. Because antibodies comprise two heavy chain / light chain pairs, a Cys-engineered antibody comprising the unpaired Cys residue at position 1 8 of the light chain (Eu numbering) will comprise two unpaired Cys per antibody, a Cys-engineered antibody comprising the unpairedCys residue at position 375 of the heavy chain constant domain (Eu numbenng) will comprise two unpaired Cys per antibody, and a Cys-engineered antibody comprising the unpaired Cys residue at position 168 of the light chain (Eu numbering) and an unpaired Cys residue at position 375 of the heavy chain constant domain (Eu numbering) will comprise four unpaired Cys per antibody. The Cys-engineered antibody can be of various isotypes, for example, IgGl. IgG2, IgG3, or IgG4.

[0100] In the description of the present invention, the presence of an engineered cysteine, for example at heavy chain position 375 of an antibody (Eu numbering) may be abbreviated as 375 C or S375C since the serine at position 375 is replaced with cysteine, at light chain position 168 of an antibody (Eu numbering) may be abbreviated as 168C or S168C since the serine at position 1 8 is replaced with cysteine, etc.

[0101] As used herein, the term "Fc domain'’, or “Fc” as used herein is the crystallizable fragment domain or region obtained from an antibody that comprises the CH2 and CH3 domains of an antibody. In an antibody, the two Fc domains are held together by two or more disulfide bonds and by hydrophobic interactions of the CH3 domains. The Fc domain may be obtained by digesting an antibody with the protease papain. Typically, amino acids in the Fc domain are numbered according to the Eu numbering convention (See Edelmann et al.. Biochem. 63: 78-85 (1969)).

[0102] The term "antibody-drug conjugate", “antibody conjugate” or "ADC" is an antibody or binder that is conjugated to one or more (typically from 1 to 8) pay loads, each through a linker to a specific site on the antibody or binder. The “drug” component of an ADC as used herein may refer to a cytotoxic agent, an anti-inflammatory agent, a radionuclide, an enzyme, a biologically active small molecule or peptide, a detectable compound, or an imaging agent. In particular embodiments, the antibody is typically a monoclonal antibody that specifically binds a tumor- associated antigen (TAA) and is capable of delivering the pay load conjugated thereto into a cell expressing the TAA on the extracellular surface of the cell. In other embodiments, the antibody is a monoclonal antibody that is conjugated to a payload for the purpose of enhancing the pharmacokinetic properties of the payload.

[0103] The terms “drug load” or “drug loading” refer to the molar ratio of drug molecules per antibody in an individual ADC molecule. In embodiments disclosed herein, the drug loading may comprise from 1-4 drug molecules, from 2-4 drug molecules, from 1-3 drug molecules, or from 2-3 drug molecules. In embodiments, the drug loading may comprise 1 drug molecule, 2 drug molecules, 3 drug molecules, or 4 drug molecules. In embodiments, the general formula of an ADC molecule may be represented as A(-L-D)n, wherein A is the antibody, L is the linker, D isthe drug, and n is the number of drug molecules linked to the antibody. In embodiments, n is from 1-4, or from 2-4, or from 1-3, or from 2-3. In embodiments, n is 1, 2, 3, or 4. As used herein, a particular drug-loading species is referred to as “DAR-n” or “D-n”, where n is the drug loading of the species (for example, a species with 2 drug molecules attached to the antibody would be DAR2 or D2, a species with 4 drug molecules attached to the antibody would be DAR4 or D4, and so forth).

[0104] The terms “drug-to-antibody ratio” or “DAR” of a given composition refer to the weighted average molar ratio of drug molecules per antibody in a population of at least two ADC molecules. Despite the conjugate specificity provided by the methods of the present disclosure, a given population of ADCs may comprise ADC molecules having different drug loadings (e.g., ranging from 1 to 4 in the case of a Cys-engineered antibody). That is, following conjugation, ADC compositions of the invention may comprise a mixture of ADCs with different drug loadings. Hence, DAR represents the weighted average of drug loadings for the ADC population as a whole (i.e. , all the ADC molecules taken together). The term "‘composition” as used above, is understood to encompass pharmaceutical compositions. Average DAR can be determined by various conventional means such as UV spectroscopy, mass spectroscopy, ELISA assay, radiometric methods, hydrophobic interaction chromatography (HIC), electrophoresis and HPLC.

[0105] In embodiments, the ADCs of the present disclosure have a DAR of about 2, or a DAR of about 4. In this context, the term ‘‘about” means an amount within ±7.5% of the actual value; i.e., “about 2” means 1.85, 1.86, 1.87, 1.88, 1.89, 1.90, 1.91, 1.92, 1.93, 1.94, 1.95, 1.96, 1.97, 1.98, 1.99, 2.00, 2.01, 2.02, 2.03, 2.04, 2.05, 2.06, 2.07, 2.08, 2.09, 2.10, 2.11, 2.12, 2.13, 2.14, 2. 15, and any number or range between them.Design Principles

[0106] The disclosure provides design principles that may be widely applied in identifying or designing phosphine-based reducing agents capable of selectively reducing any engineered cysteine site on the antibody backbone. To do so, a library of commercially relevant phenylphosphines was screened to assess their ability to selectively reduce engineered cysteine sites found on both the Fc (S375C) and Fab (S168C) regions of multiple antibodies. By incorporating in silico modeling (Fig. IB) to determine accessibility scores and studying the inherent reduction kinetics, coupled with analytical characterization of ADCs, insights were made into why certain phosphines exhibited selective reduction of engineered cysteine sites. The approach exhibits remarkable versatility. By identifying reductants with distinct reactivity capable of selectively reducing specific engineered cysteines, dual-payload ADCs wereconstructed without having to use complex conjugation chemistry or multifunctional branched linkers. This advancement streamlines the manufacturing process of site-specific ADCs and enables the use of entirely new avenues for optimizing their therapeutic properties. The implications of the methods disclosed herein are far-reaching, as it unlocks the development of highly customizable and potent ADCs with enhanced efficacy and reduced toxicity.Exemplary embodiments of the present invention

[0107] The present invention provides a method for selectively reducing a capped engineered cysteine residue over interchain disulfide bonds of a Cys-engineered antibody to produce a reduced, uncapped Cys-engineered antibody, the method comprising contacting the Cys- engineered antibody with a phosphine-based reductant for a time sufficient to reduce the capped engineered cysteine of the Cys-engineered antibody to produce the reduced, uncapped Cys- engineered antibody.

[0108] In particular embodiments, the phosphine-based reductant is selected from the group consisting of: 4-(diphenylphosphaneyl)benzenesulfonic acid (2), 4,4'- (phenylphosphanediyl)dibenzenesulfonic acid (3), 3-(diphenylphosphaneyl)benzenesulfonic acid (4), 4-(diphenylphosphaneyl)benzoic acid (5), 4-(diphenylphosphaneyl)-N,N-dimethylaniline (6), (R)-2-((diphenylphosphaneyl)methyl)pyrrolidine (7), 2-(diphenylphosphaneyl)benzenesulfonic acid (8), triphenylphosphine (9), 2-(diphenylphosphaneyl)benzaldehyde (10), tris(4- methoxyphenyljphosphane (11), 2-(di-p-tolylphosphaneyl)benzaldehyde (12), 2- (dicyclohexylphosphaneyl)benzenesulfonic acid (13), 4-(bis(4- fluorophenyl)phosphaneyl)benzenesulfonic acid (24), tris(4-fluorophenyl)phosphane (28), 4-(diphenylphosphaneyl)-N.N,N-trimethylbenzenaminium (29), 4-((4- fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30), and salts thereof.

[0109] In particular embodiments, the phosphine-based reductant is 4- (diphenylphosphaneyl)benzenesulfonic acid (2) or salt thereof.

[0110] In particular embodiments, the phosphine-based reductant is 4,4'- (phenylphosphanediyl)dibenzenesulfonic acid (3) or salt thereof.[OHl] In particular embodiments, the phosphine-based reductant is 3- (diphenylphosphaneyl)benzenesulfonic acid (4) or salt thereof.

[0112] In particular embodiments, the phosphine-based reductant is 4- (diphenylphosphaneyl)benzoic acid (5) or salt thereof.

[0113] In particular embodiments, the phosphine-based reductant is 4-(diphenylphosphaneyl)- N,N-dimethylaniline (6) or salt thereof.

[0114] In particular embodiments, the phosphine-based reductant is (R)-2- ((diphenylphosphaneyl)methyl)pyrrolidine (7) or salt thereof.

[0115] In particular embodiments, the phosphine-based reductant is 2- (diphenylphosphaneyl)benzenesulfonic acid (8) or salt thereof.

[0116] In particular embodiments, the phosphine-based reductant is triphenylphosphine (9), 2- (diphenylphosphaneyl)benzaldehyde (10) or salt thereof.

[0117] In particular embodiments, the phosphine-based reductant is tris(4- methoxyphenyl)phosphane (11) or salt thereof.

[0118] In particular embodiments, the phosphine-based reductant is 2-(di-p- tolylphosphaneyl)benzaldehyde (12) or salt thereof.

[0119] In particular embodiments, the phosphine-based reductant is 2- (dicyclohexylphosphaneyl)benzenesulfonic acid (13) or salt thereof.

[0120] In particular embodiments, the phosphine-based reductant is 4-(bis(4- fluorophenyl)phosphaneyl)benzenesulfonic acid (24) or salt thereof.

[0121] tris(4-fluorophenyl)phosphane (28) or salt thereof.

[0122] In particular embodiments, the phosphine-based reductant is 4-(diphenylphosphaneyl)- N,N,N-trimethyIbenzenaminium (29) or salt thereof.

[0123] In particular embodiments, the phosphine-based reductant is 4-((4- fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30) or salt thereof.

[0124] In a further embodiment of the method, the Cys-engineered antibody comprises one or more engineered cysteine residues, wherein each engineered cysteine residue is located at a position within the Cys-engineered antibody selected from the group consisting of heavy chain positions 152, 153. 171, 172, 173. 347, 362. 375, 380. 382. 393, 430. and 440; wherein the position number is according to EU numbering; and / or the group consisting of heavy chain positions 40, 41, and 89; wherein the position number is according to Kabat numbering.

[0125] In a further embodiment of the method, the Cys-engineered antibody comprises one or more engineered cysteine residues, wherein each engineered cysteine residue is located at a position within the Cys-engineered antibody selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering; and / or the group consisting of light chain positions 40 and 41 ; wherein the position number is according to Kabat numbering.

[0126] In a further embodiment of the method, the Cys-engineered antibody comprises (a) a first engineered cysteine residue located at a position selected from the group consisting of heavy chain positions 152, 153, 171, 172, 173, 347, 362, 375, 380, 382, 393, 430, and 440; wherein theposition number is according to EU numbering; and (b) a second engineered cysteine residue located at a position selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering.

[0127] In a further embodiment of the method, the Cys-engineered antibody comprises (a) a first engineered cysteine residue at heavy chain position 375; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue at light chain position 168; wherein the position number is according to EU numbering.

[0128] In a further embodiment of the method, the reduced, uncapped Cys-engineered antibody is separated from the reductant and the reduced, uncapped Cys-engineered antibody is incubated with a payload or linker-payload comprising a thiol reactive group for a time sufficient to produce a conjugate comprising the Cys-engineered antibody conjugated to the payload or linkerpayload.

[0129] The present invention further provides a method for producing a composition of antibody conjugates, the method comprising: (a) providing a composition of Cys-engineered antibodies, wherein each antibody comprises (i) two heavy chains and tw o light chains, (ii) one or more capped engineered cysteine residues, and (iii) interchain disulfide bonds between the hinge region of the two heavy chains and betw een the hinge region of the heavy chain and the light chain; (b) mixing the composition with an excess of a phosphine-based reductant or salt thereof to provide a reaction mixture; (c) incubating the reaction mixture for a time sufficient to selectively reduce the capped engineered cysteines of the Cys-engineered antibodies over the interchain disulfide bonds of the Cys-engineered antibodies to produce reduced, uncapped Cys- engineered antibodies; (d) separating the reduced, uncapped Cys-engineered antibodies from the reductant to provide a composition of separated, reduced, uncapped Cys-engineered antibodies; and (e) incubating the composition of separated, reduced, uncapped Cys-engineered antibodies with an excess of payload comprising a thiol reactive group for a time sufficient to produce the composition of antibody conjugates.

[0130] In particular embodiments, the phosphine-based reductant is selected from the group consisting of: 4-(diphenylphosphaneyl)benzenesulfonic acid (2), 4,4'- (phenylphosphanediyl)dibenzenesulfonic acid (3), 3-(diphenylphosphaneyl)benzenesulfonic acid (4), 4-(diphenylphosphaneyl)benzoic acid (5), 4-(diphenylphosphaneyl)-N,N-dimethylaniline (6), (R)-2-((diphenylphosphaneyl)methyl)pyrrolidine (7), 2-(diphenylphosphaneyl)benzenesulfonic acid (8), triphenylphosphine (9), 2-(diphenylphosphaneyl)benzaldehyde (10), tris(4- methoxyphenyljphosphane (11), 2-(di-p-tolylphosphaneyl)benzaldehyde (12), 2-(dicyclohexylphosphaneyl)benzenesulfonic acid (13). 4-(bis(4- fluorophenyl)phosphaneyl)benzenesulfonic acid (24), tris(4-fluorophenyl)phosphane (28), 4-(diphenylphosphaneyl)-N,N,N-trimethylbenzenaminium (29), 4-((4- fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30), and salts thereof.

[0131] In particular embodiments, the phosphine-based reductant is 4- (diphenylphosphaneyl)benzenesulfonic acid (2) or salt thereof.

[0132] In particular embodiments, the phosphine-based reductant is 4,4'- (phenylphosphanediyl)dibenzenesulfonic acid (3) or salt thereof.

[0133] In particular embodiments, the phosphine-based reductant is 3- (diphenylphosphaneyl)benzenesulfonic acid (4) or salt thereof.

[0134] In particular embodiments, the phosphine-based reductant is 4- (diphenylphosphaneyl)benzoic acid (5) or salt thereof.

[0135] In particular embodiments, the phosphine-based reductant is 4-(diphenylphosphaneyl)- N,N-dimethylaniline (6) or salt thereof.

[0136] In particular embodiments, the phosphine-based reductant is (R)-2- ((diphenylphosphaneyl)methyl)pyrrolidine (7) or salt thereof.

[0137] In particular embodiments, the phosphine-based reductant is 2- (diphenylphosphaneyl)benzenesulfonic acid (8) or salt thereof.

[0138] In particular embodiments, the phosphine-based reductant is triphenylphosphine (9), 2- (diphenylphosphaneyl)benzaldehyde (10) or salt thereof.

[0139] In particular embodiments, the phosphine-based reductant is tris(4- methoxyphenyl)phosphane (11) or salt thereof.

[0140] In particular embodiments, the phosphine-based reductant is 2-(di-p- tolylphosphaneyl)benzaldehyde (12) or salt thereof.

[0141] In particular embodiments, the phosphine-based reductant is 2- (dicyclohexylphosphaneyl)benzenesulfonic acid (13) or salt thereof.

[0142] In particular embodiments, the phosphine-based reductant is 4-(bis(4- fluorophenyl)phosphaneyl)benzenesulfonic acid (24) or salt thereof.

[0143] tris(4-fluorophenyl)phosphane (28) or salt thereof.

[0144] In particular embodiments, the phosphine-based reductant is 4-(diphenylphosphaneyl)- N,N,N-trimethyIbenzenaminium (29) or salt thereof.

[0145] In particular embodiments, the phosphine-based reductant is 4-((4- fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30) or salt thereof.

[0146] In a further embodiment of the method, the Cys-engineered antibodies comprise one or more engineered cysteine residues, wherein each engineered cysteine residue is located at a position within the Cys-engineered antibody selected from the group consisting of heavy chain positions 152, 153. 171, 172, 173, 347, 362, 375, 380. 382, 393, 430, and 440; wherein the position number is according to EU numbering; and / or the group consisting of heavy chain positions 40, 41, and 89; wherein the position number is according to Kabat numbering.

[0147] In a further embodiment of the method, Cys-engineered antibodies comprise one or more engineered cysteine residues, wherein each engineered cysteine residue is located at a position within the Cys-engineered antibody selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering; and / or the group consisting of light chain positions 40 and 41; wherein the position number is according to Kabat numbering.

[0148] In a further embodiment of the method, the Cys-engineered antibodies comprise (a) a first engineered cysteine residue located at a position selected from the group consisting of heavy chain positions 152, 153, 171, 172, 173, 347, 362, 375, 380, 382, 393, 430, and 440; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue located at a position selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering.

[0149] In a further embodiment of the method, the Cys-engineered antibodies comprise (a) a first engineered cysteine residue at heavy chain position 375; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue at light chain position 168; wherein the position number is according to EU numbering.

[0150] In a further embodiment of the method, the excess of reductant comprises 6 equivalences of reductant per engineered cysteine.

[0151] In a further embodiment of the method, the excess of pay load comprises 4 equivalences of payload per engineered cysteine.

[0152] In a further embodiment of the method, greater than 90% of the Cys-engineered antibodies in the composition of antibody conjugates are intact following reduction and conjugation.

[0153] In a further embodiment of the method, the composition of antibody conjugates has a drug-to-antibody ratio (DAR) between 1.7 and 2.0.

[0154] The present invention further provides a method for producing a composition of Cys- engineered antibodies having a tetrameric structure in which the engineered cysteine residues thereof are selectively reduced over interchain disulfide bonds of the Cys-engineered antibodies,the method comprising (a) providing a composition of Cys-engineered antibodies comprising capped engineered cysteines and a phosphine-based reductant or salt thereof; (b) mixing the composition with the reductant to provide a reaction mixture; and (c) incubating the reaction mixture for a time sufficient to reduce the capped engineered cysteines of the Cys-engineered antibody to produce the reduced, uncapped Cys-engineered antibody.

[0155] In particular embodiments, the phosphine-based reductant is selected from the group consisting of: 4-(diphenylphosphaneyl)benzenesulfonic acid (2), 4,4'- (phenylphosphanediyl)dibenzenesulfonic acid (3). 3-(diphenylphosphaneyl)benzenesulfonic acid (4). 4-(diphenylphosphaneyl)benzoic acid (5). 4-(diphenylphosphaneyl)-N.N-dimethylaniline (6), (R)-2-((diphenylphosphaneyl)methyl)pyrrolidine (7), 2-(diphenylphosphaneyl)benzenesulfonic acid (8), triphenylphosphine (9), 2-(diphenylphosphaneyl)benzaldehyde (10), tris(4- methoxyphenyl)phosphane (11), 2-(di-p-tolylphosphaneyl)benzaldehyde (12), 2- (dicyclohexylphosphaneyl)benzenesulfonic acid (13), 4-(bis(4- fluorophenyl)phosphaneyl)benzenesulfonic acid (24). tris(4-fluorophenyl)phosphane (28).4-(diphenylphosphaneyl)-N,N,N-trimethylbenzenaminium (29), 4-((4- fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30), and salts thereof.

[0156] In particular embodiments, the phosphine-based reductant is 4- (diphenylphosphaneyl)benzenesulfonic acid (2) or salt thereof.

[0157] In particular embodiments, the phosphine-based reductant is 4,4'- (phenylphosphanediyl)dibenzenesulfonic acid (3) or salt thereof.

[0158] In particular embodiments, the phosphine-based reductant is 3- (diphenylphosphaneyl)benzenesulfonic acid (4) or salt thereof.

[0159] In particular embodiments, the phosphine-based reductant is 4- (diphenylphosphaneyl)benzoic acid (5) or salt thereof.

[0160] In particular embodiments, the phosphine-based reductant is 4-(diphenylphosphaneyl)- N,N-dimethylaniline (6) or salt thereof.

[0161] In particular embodiments, the phosphine-based reductant is (R)-2- ((diphenylphosphaneyl)methyl)pyrrolidine (7) or salt thereof.

[0162] In particular embodiments, the phosphine-based reductant is 2- (diphenylphosphaneyl)benzenesulfonic acid (8) or salt thereof.

[0163] In particular embodiments, the phosphine-based reductant is triphenylphosphine (9), 2- (diphenylphosphaneyl)benzaldehyde (10) or salt thereof.

[0164] In particular embodiments, the phosphine-based reductant is tris(4- methoxyphenyl)phosphane (11) or salt thereof.

[0165] In particular embodiments, the phosphine-based reductant is 2-(di-p- tolylphosphaneyl)benzaldehyde (12) or salt thereof.

[0166] In particular embodiments, the phosphine-based reductant is 2- (dicyclohexylphosphaneyl)benzenesulfonic acid (13) or salt thereof.

[0167] In particular embodiments, the phosphine-based reductant is 4-(bis(4- fluorophenyl)phosphaneyl)benzenesulfonic acid (24) or salt thereof.

[0168] tris(4-fluorophenyl)phosphane (28) or salt thereof.

[0169] In particular embodiments, the phosphine-based reductant is 4-(diphenylphosphaneyl)- N,N,N-trimethylbenzenaminium (29) or salt thereof.

[0170] In particular embodiments, the phosphine-based reductant is 4-((4- fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30) or salt thereof.

[0171] In a further embodiment of the method, the Cys-engineered antibodies comprise one or more engineered cysteine residues, wherein each engineered cysteine residue is located at a position within the Cys-engineered antibody selected from the group consisting of heavy chain positions 152, 153, 171, 172, 173, 347, 362, 375, 380, 382, 393, 430, and 440; wherein the position number is according to EU numbering; and / or the group consisting of heavy chain positions 40, 41, and 89; wherein the position number is according to Kabat numbering.

[0172] In a further embodiment of the method, Cys-engineered antibodies comprise one or more engineered cysteine residues, wherein each engineered cysteine residue is located at a position within the Cys-engineered antibody selected from the group consisting of light chain positions 1 5 and 168; wherein the position number is according to EU numbering; and / or the group consisting of light chain positions 40 and 41; wherein the position number is according to Kabat numbering.

[0173] In a further embodiment of the method, the Cys-engineered antibodies comprise (a) a first engineered cysteine residue located at a position selected from the group consisting of heavy chain positions 152, 153, 171, 172, 173, 347, 362. 375, 380, 382, 393, 430, and 440; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue located at a position selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering.

[0174] In a further embodiment of the method, the Cys-engineered antibodies comprise (a) a first engineered cysteine residue at heavy chain position 375; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue at light chain position 168; wherein the position number is according to EU numbering.

[0175] In a further embodiment of the method, greater than 90% of the reduced, uncapped Cys- engineered antibodies in the composition are intact.

[0176] The present invention further provides a method for producing a composition of antibody conjugates conjugated to a first and a second pay load, the method comprising: (a) providing a mixture comprising Cys-engineered antibodies having a heavy chain and a light chain, a first capped engineered cysteine residue at position 168 of the light chain, a second capped engineered cysteine residue at position 375 of the heavy chain, and interchain disulfide bonds, wherein the position numbers are according to EU numbering; (b) adding to the mixture an excess of first reductant 2-(diphenylphosphaneyl)benzenesulfonic acid (8) or salt thereof to produce a first reducing mixture, and incubating the first reducing mixture for a time sufficient to selectively reduce the first capped engineered cysteines at position 168 of the light chains without reducing the interchain disulfide bonds of the Cys-engineered antibodies; (c) separating the first reduced, uncapped Cys-engineered antibodies from the first reductant to provide a first composition of Cys-engineered antibodies; (d) adding an excess of a first payload comprising a thiol reactive group to the first composition to produce a first conjugation mixture and incubating the first conjugation mixture for a time sufficient to produce Cys-engineered antibodies conjugated to the first payload; (e) separating the Cys-engineered antibodies conjugated to the first payload from the first conjugation mixture to provide a composition of Cys-engineered antibodies conjugated to the first payload; (f) adding to the composition of Cys-engineered antibodies conjugated to the first payload an excess of a second reductant 4- (diphenylphosphaneyl)benzenesulfonic acid (2) or salt thereof to produce a second reducing mixture, and incubating the second reducing mixture for a time sufficient to selectively reduce the second capped engineered cysteines at position 375 of the heavy chains without reducing the interchain disulfide bonds of the Cys-engineered antibodies to produce second reduced, uncapped Cys-engineered antibodies; (g) separating the second reduced, uncapped Cys-engineered antibodies from the second reductant to provide a second composition of Cys-engineered antibodies; (h) adding an excess of a second payload comprising a thiol reactive group to the second composition to produce a second conjugation mixture and incubating the second conjugation mixture for a time sufficient to produce Cys-engineered antibodies conjugated to the second payload; and (i) separating the Cys-engineered antibodies conjugated to the second payload from the second conjugation mixture to provide a composition of Cys-engineered antibodies conjugated to the first payload and the second payload.

[0177] In a further embodiment of the method, the excess of the first reductant comprises about 40 equivalences of reductant per capped engineered cysteine.

[0178] In a further embodiment of the method, the excess of the second reductant comprises about 6 equivalences of reductant per capped engineered cysteine.

[0179] In a further embodiment of the method, the excess of pay load comprises 4 equivalences of payload per reduced engineered cysteine.

[0180] The present invention further provides a dual-payload antibody conjugate comprising an antibody having a light chain comprising a cysteine residue at position 168 conjugated to a first payload and a heavy chain comprising a cysteine residue at position 375 conjugated to a second payload, wherein the position numbering is according to Eu numbering and wherein the first payload and the second payload are not the same.

[0181] The present invention further provides an antibody conjugate comprising an antibody having a light chain and a heavy chain, wherein the light chain comprises an engineered cysteine at position 168 conjugated to a first payload and the heavy' chain comprises an engineered cysteine residue at position 375 conjugated to a second payload, and wherein the position numbering is according to Eu numbering and wherein the first payload and the second payload are not the same.

[0182] The present invention further provides a reductant selected from the group consisting of: 4-(bis(4-fluorophenyl)phosphaneyl)benzenesulfonic acid (24), 4-(diphenylphosphaneyl)-N,N,N-trimethylbenzenaminium (29), 4-((4- fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30), and salts thereof.

[0183] The present invention further provides a reductant selected from the group consisting of: 4-(dicyclohexylphosphaneyl)benzoic acid (22), 4-(dicyclohexylphosphaneyl)benzenesulfonic acid (23), 4-(bis(4-methoxyphenyl)phosphaneyl)benzenesulfonic acid (25). 4- (cyclohexyl(phenyl)phosphaneyl)benzoic acid (26). 4- (cyclohexyl(phenyl)phosphaneyl)benzenesulfonic acid (27), and salts thereof.

[0184] The present invention further provides a method for selectively reducing capped engineered cysteine residues over interchain disulfide bonds of an antibody comprising incubating the antibody with an excess of a phosphine-based reductant for a time sufficient to reduce the capped engineered cysteine residues, wherein the phosphine-based reductant is 4- (bis(4-fluorophenyl)phosphaneyl)benzenesulfonic acid (24) or salt thereof.

[0185] The present invention further provides a method for selectively reducing capped engineered cysteine residues over interchain disulfide bonds of an antibody comprising incubating the antibody with an excess of a phosphine-based reductant for a time sufficient to reduce the capped engineered cysteine residues, wherein the phosphine-based reductant is 4- (diphenylphosphaneyl)-N,N,N-trimethylbenzenaminium (29) or salt thereof.

[0186] The present invention further provides a method for selectively reducing capped engineered cysteine residues over interchain disulfide bonds of an antibody comprising incubating the antibody with an excess of a phosphine-based reductant for a time sufficient to reduce the capped engineered cysteine residues, wherein the phosphine-based reductant is 4-((4- fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30) or salt thereof.

[0187] The present invention further provides a method for reducing capped engineered cysteine residues over interchain disulfide bonds of an antibody comprising incubating the antibody with an excess of a phosphine-based reductant for a time sufficient to reduce the capped engineered cysteine residues, wherein the phosphine-based reductant is selected from the group consisting of: 4-(dicyclohexylphosphaneyl)benzoic acid (22), 4- (dicyclohexylphosphaneyl)benzenesulfonic acid (23), 4-(bis(4- methoxyphenyl)phosphaneyl)benzenesulfonic acid (25), 4- (cyclohexyl(phenyl)phosphaneyl)benzoic acid (26), 4- (cyclohexyl(phenyl)phosphaneyl)benzenesulfonic acid (27), and salts thereof.

[0188] The present invention further provides a method for reducing capped engineered cysteine residues over interchain disulfide bonds of an antibody comprising incubating the antibody with an excess of a phosphine-based reductant for a time sufficient to reduce the capped engineered cysteine residues, wherein the phosphine-based reductant is 4- (dicyclohexylphosphaneyl)benzoic acid (22) or salt thereof.

[0189] The present invention further provides a method for reducing capped engineered cysteine residues over interchain disulfide bonds of an antibody comprising incubating the antibody with an excess of a phosphine-based reductant for a time sufficient to reduce the capped engineered cysteine residues, wherein the phosphine-based reductant is 4- (dicyclohexylphosphaneyl)benzenesulfonic acid (23) or salt thereof.

[0190] The present invention further provides a method for reducing capped engineered cysteine residues over interchain disulfide bonds of an antibody comprising incubating the antibody with an excess of a phosphine-based reductant for a time sufficient to reduce the capped engineered cysteine residues, wherein the phosphine-based reductant is 4-(bis(4- methoxyphenyl)phosphaneyl)benzenesulfonic acid (25) or salt thereof.

[0191] The present invention further provides a method for reducing capped engineered cysteine residues over interchain disulfide bonds of an antibody comprising incubating the antibody with an excess of a phosphine-based reductant for a time sufficient to reduce the capped engineered cysteine residues, wherein the phosphine-based reductant is 4- (cyclohexyl(phenyl)phosphaneyl)benzoic acid (26) or salt thereof.

[0192] The present invention further provides a method for reducing capped engineered cysteine residues over interchain disulfide bonds of an antibody comprising incubating the antibody with an excess of a phosphine-based reductant for a time sufficient to reduce the capped engineered cysteine residues, wherein the phosphine-based reductant is 4- (cyclohexyl(phenyl)phosphaneyl)benzenesulfonic acid (27) or salt thereof.

[0193] The present invention further provides a method for reducing capped engineered cysteine residues over interchain disulfide bonds of an antibody comprising incubating the antibody with an excess of a phosphine-based reductant for a time sufficient to reduce the capped engineered cysteine residues, wherein the phosphine-based reductant is tris(4- fluorophenyl)phosphane (28) or salt thereof.Payload

[0194] The payload may be a therapeutic agent, a detectable label or moiety, radionuclide, or protecting group. In particular embodiments, the payload may possess a desired biological activity. The desired biological activity includes the diagnosis, cure, mitigation, treatment, or prevention of disease in an animal such as a human. The term "payload" refers to chemicals recognized as drugs in the official United States Pharmacopeia, official Homeopathic Pharmacopeia of the United States, or official National Formulary, or any supplement thereof. Exemplary drugs are set forth in the Physician's Desk Reference (PDR) and in the Orange Book maintained by the U.S. Food and Drug Administration (FDA). New drugs are being continually being discovered and developed, and the present invention provides that these new drugs may also be included in the term payload.

[0195] The payload can be linked to the antibody or fragment by a linker. Suitable linkers include, for example, cleavable and non-cleavable linkers. A cleavable linker is typically susceptible to cleavage under intracellular conditions. Suitable cleavable linkers include, for example, a peptide linker cleavable by an intracellular protease, such as lysosomal protease or an endosomal protease. In exemplary embodiments, the linker can be a dipeptide linker, such as a valine-citrulline (val-cit), a phenylalanine-lysine (phe-lys) linker, or maleimidocapronic-valine- citruline-p-aminobenzyloxy carbonyl (MC-Val-Cit-PABC) linker. Another linker is Sulfosuccinimidyl-4-[N-maleimidomethyl]cyclohexane-l-carboxylate (smcc). Sulfo-smcc conjugation occurs via a maleimide group which reacts with sulfhydryls (thiols, -SH), while its Sulfo-NHS ester is reactive toward primary amines (as found in Lysine and the protein or peptide N-terminus). Yet another linker is maleimidocaproyl (me). Other suitable linkers include linkers hydrolyzable at a specific pH or a pH range, such as a hydrazone linker. Additional suitablecleavable linkers include disulfide linkers. The linker may be covalently bound to the antibody to such an extent that the antibody must be degraded intracellularly in order for the drug to be released e.g., the me linker and the like.

[0196] The pay load-antibody conjugate is effective for the usual purposes for which the corresponding drugs are effective, but have superior efficacy because of the ability, inherent in the antibody, to transport the payload to the desired cell where it is of particular benefit. Exemplar}7payloads include proteins, peptides, and small molecules. More specifically, these payloads include, for example, enzyme inhibitors such as dihydrofolate reductase inhibitors, and thymidylate synthase inhibitors, DNA intercalators, glutocorticoid receptor agonists, nuclear receptor agonists, antinflammatory agents, DNA cleavers, degraders, topoisomerase inhibitors, anthracycline drugs, vinca drugs, mitomycins, bleomycins, cytotoxic nucleosides, pteridine drugs, diynenes, podophyllotoxins, differentiation inducers, and taxols.

[0197] In one embodiment, the payload may be a cytotoxic drug useful in cancer therapy and other small molecules, proteins or polypeptides with desired biological activity, such as a toxin. The payload may be selected to be activated at a tumor cells by conjugation to a tumor-specific antibody. These tumor specific payload-antibody conjugates have tumor specificity arising from the specificity of the antibody. Examples of this are payload-antibody conjugates that are highly selective substrates for tumor specific enzymes, where these enzymes are present in the proximity of the tumor in sufficient amounts to generate cytotoxic levels of free payload in the vicinity7of the tumor.

[0198] Cytotoxic drugs include, for example, duocarmycins and CC-1065, and analogues thereof, including CBI (l,2,9,9a-tetrahydrocyclopropa[c]benz[e]indol-4-one)-based analogues, MCBI (7-methoxy-l,2,9,9a-tetra-hydrocyclopropa[c]benz[e]indol-4-one)-based analogues and CCBI (7-cyano-l,2,9,9a-tetra-hydrocyclo-propa[c]benz[e]-indol-4-one)-based analogues of the duocarmycins and CC-1065, doxorubicin and doxorubicin conjugates such as morpholinodoxorubicin and cy anomorpholino-doxorubicin, dolastatins such as dolestatin- 10, combretastatin, calicheamicin, maytansine, maytansine analogs, maytansinoid derivatives DM-1. DM-2. DM-4, auristatin E, auristatin EB (AEB), auristatin EFP (AEFP), methyl auristatin E (MMAE), monomethyl auristatin F (MMAF), 5-benzoylvaleric acid- AE ester (AEVB), tubulysins, disorazole, epothilones, Paclitaxel, docetaxel, SN-38, Topotecan, rhizoxin, echinomycin. colchicine, vinblastin, vindesine, estramustine, cemadotin, eleutherobin, methotrexate, methopterin, dichloro methotrexate, 5-fluorouracil, 6-mercaptopurine, cytosine arabinoside, melphalan, leurosine, leurosideine, actinomycin, daunorubicin and daunorubicin conjugates, mitomycin C, mitomycin A, carminomycin, aminopterin, tallysomycin, podophyllotoxin andpodophyllotoxin derivatives such as etoposide or etoposide phosphate, vincristine, taxol. taxotere retinoic acid, butyric acid, N-acetyl spermidine, camptothecin, PNU, and their analogues.

[0199] In particular embodiments, the payload is an anti-inflammatory agent, for example, a glucocorticoid receptor agonist. Representative glucocorticoid receptor agonists include but are not limited to glucocorticoids such as Cortisol, cortisone acetate, beclometasone. prednisone, prednisolone, methylprednisolone, betamethasone, trimcinolone, budesonide, dexamethasone, fluticasone, fluticasone propionate, fluticasone furoate, or mometasone.EXAMPLES

[0200] The following examples are meant to be illustrative and should not be construed as further limiting. The contents of the figures and all references, patents, and published patent applications cited throughout this application are expressly incorporated herein by reference.1) Materials and methods

[0201] The reductants Triphenylphosphine-3,3 ',3 ''-trisulfonic acid trisodium salt (1), 4- (diphenylphosphaneyl)benzenesulfonic acid (2), 4,4'-(phenylphosphanediyl)dibenzenesulfonic acid (3), 3-(diphenylphosphaneyl)benzenesulfonic acid (4), 4-(diphenylphosphaneyl)benzoic acid (5). 4-(diphenylphosphaneyl)-N.N-dimethylaniline (6), (R)-2- ((diphenylphosphaneyl)methyl)pyrrolidine (7), 2-(diphenylphosphaneyl)benzenesulfonic acid (8), triphenylphosphine (9), 2-(diphenylphosphaneyl)benzaldehyde (10), tris(4- methoxyphenyl)phosphane (11), 2-(di-p-tolylphosphaneyl)benzaldehyde (12), 2- (dicyclohexylphosphaneyl)benzenesulfonic acid (13), 2-(diphenylphosphaneyl)benzoic acid (14), 2-(diphenylphosphaneyl)ethan-l -amine (15), 3-(diphenylphosphaneyl)propan-l-amme (16). 2-(2- (diphenylphosphaneyl)ethyl)pyridine (17), and tris(2-carboxyethyl)phosphine (TCEP; 18), L- cysteine methylester (20), tris(4-fluorophenyl)phosphane (24), and tris(4- fluorophenyl)phosphane (28) were purchased from Sigma-Aldrich® (Sigma- Aldrich, Inc. St. Louis, MO USA) and were used as 10 mM stock solutions either in water or dimethylsulfoxide (DMSO) or dimethylacetamide (DMA) depending upon solubility. These stock solutions were made fresh just before starting the reduction reaction. Reductants 22-27 and 29-30 were synthesized as set forth in the Examples 1-8. Reductants 1-18, 20, 22-27, and 29-30 are show n in Table 1. All the other chemical reagents used for the synthesis of reductants 22-27 and 29-30 were used as received without further purification. MC-val-cit-PABC-monomethyl auristatin E (19) (See U.S. Pat. No. 7,964,566) and MC-val-cit-PABC-monomethyl auristatin F (21) (See U.S. Pat. No. US7,994,135) were purchased from MedChemExpress (Monmouth Junction, NJUSA) and used as 10 mM stock solutions in DMSO, wherein MC is maleimidocaprol, val is valine, cit is citrulline, PABC is p-aminobenzylcarbamate. FabRICATOR® (IdeS) enzyme (2000 units) was purchased from Genovis Inc. (Cambridge, MA USA) to confirm selective conjugations at desired sites.

[0202] All the antibody reductions and conjugations were run in the Eppendorf® ThermoMixer® Fl.5 Model 5384 with varying Eppendorf® tube sizes (0.5 ml to 15 mL tubes) (Eppendorf North America, Enfield, CT USA). Purification of antibody-drug conjugates (ADCs)was accomplished using a Cytiva™. G-25 desalting columns (PD-10 columns of different sizes / volume) (Cytiva Life Sciences, Marlborough, MA USA). The concentration and yield of the ADCs after purification were quantified based on UV-Visible light measurements on a Tecan Infinite® M Plex instrument (Mannedorf, CH) at 280 nm, correcting for the absorbance of the conjugated linker-payload.

[0203] 1H NMR spectra’s were recorded on a Bruker 400 MHz instrument equipped with a 5mm iProbe.1H NMR spectra were internally referenced to the residual proteo-solvent signals (note: DMSO-de referenced at 2.50 and 39.52 ppm respectively). 'l l NMR were reported as follows: chemical shift (d ppm), multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, p = pentet, h = hextet, hept = heptet, m = multiplet, br = broad), coupling constant (Hz), and integration.

[0204] Liquid chromatography-mass spectrometry (LC-MS) analysis for triphenylphosphine ligands was performed on an Agilent (Santa Clara, CA USA) 1290 Infinity II LC system with an Infinity Lab liquid Chromatography / Mass Spectrometry Detector (LC / MSD) (Waters (Waters Corporation, Milford, MA USA) BEH Cl 8 column, 50 x 2.1 mm, 1.7 pm particle size; water / methyl cyanide (MeCN) gradient with 0.1% TFA). LC-MS analysis for antibodies and antibody-drug conjugates (ADCs) was performed an Agilent 1290 Infinity II LC system with an Agilent 6230 LC / TOF (Acquity UPLC® Peptide CSHTMC18 column, 130A, 1.7 pm for peptides; polymeric reverse-phase S (PLRP-S) column, 1000A, 5 pm for antibodies / ADCs; MeCN / water + 0.1% trifluoroacetate (TFA) mobile phase). Data were collected and analyzed by deconvolution using Agilent MassHunter BioConfirm Software 10.0. Drug-to-antibody ratio (DAR) was determined through integration of deconvoluted MS peaks.

[0205] Size-exclusion chromatography (SEC) of antibodies and ADCs was performed on an Agilent 1290 Infinity II LC system (Waters Acquity BEH200 SEC column, 4.6 x 150 mm, 1.7 pm particle size; mobile phase = 100 mM sodium phosphate, 200 mM sodium chloride, 0.04% sodium azide, 5% IP A, pH 7.0). % Purity and selective reduction of ADCs were evaluated byrunning capillary electrophoresis sodium dodecyl sulfate (CE-SDS) assay in non-reduced mode 10 pg sample size each run.2) General procedure for selective reduction ofS375C or S168C Cys-engineered IgGl antibody and conjugation

[0206] IgGl type Cys-engineered antibody comprising S168C or S375C substitutions (1 mg, 105pL from 9.53 mg / ml stock) was buffer exchanged into 100 mM Tris 10 mM ethylenediaminetetraacetic acid (EDTA) in phosphate buffered saline (PBS) 7.4 (pH 8.2) in 1.5mL Eppendorf® tube. 6 equivalences of reductant (1-30) (4 pL) from 10 mM stock solution in water,) and volume was adjusted using the same buffer to final concentration of 5 mg / rnL. The Eppendorf® tube was incubated on thermomixer at 37 °C for 2 hours under shaking conditions (800-1000 rpm). Complete reduction of engineered cysteine was confirmed by Intact mass-spec analysis demonstrating mass loss of 2 cysteines.

[0207] After reduction, the excess of the reductant was removed using Cytiva™, G-25 (PD-10 column, 0.5 mL load and 1 mL elute) column. The eluted fraction was reconcentrated to 5 mg / mL using Amicon® filter (10 kd cut off) in PBS 7.4 (200 pL final volume). This was followed by addition of 4 equivalences of linker-payload (2.7 pL of 19 from 10 mM stock) in dimethyl sulfoxide (DMSO) and the same Eppendorf® tube was incubated at 24 °C for 3 hours. Complete conjugation was confirmed by deconvolution of intact mass-spec revealing mostly DAR (drug to antibody ratio) of 2. The excess linker-payload was removed by using Cytiva™, G-25 (PD-10 column, 0.5 mL load and 1 mL elute) column. The eluted fraction was reconcentrated to 5 mg / mL using Amicon® filter (lOkd cut off) in desired stability buffer (5 mg / mL final concentration ADC).3) General protocol for stepwise reduction ofS375C andS168C Cys-engineered antibodies and conjugation of dual payloads thereto using reductants 8 and 2

[0208] Step 1 : Cys-engineered IgG type antibody comprising S168C and S375C substitutions (1 mg, 105 pL from 9.53 mg / rnL stock) was buffer exchanged into 100 mM Tris 10 mM EDTA in PBS 7.4 (pH 8.2) in 1.5 mL Eppendorf® tube. 40 equivalences of reductant 8 (27 pL from 10 mM stock solution in water,) and volume was adjusted using the same buffer to final concentration of 5 mg / mL. The Eppendorf® tube was incubated on thermomixer at 37 °C for 4 hours under shaking conditions (800-1000 revolutions per minute (rpm)). Complete reduction of S168C cysteine was confirmed by Intact mass-spec analysis demonstrating mass loss of 2 cysteines.

[0209] After reduction, the excess of the reductant was removed using Cytiva™, G-25 (PD-10 column, 0.5 mL load and 1 mL elute) column. The eluted fraction was reconcentrated to 5 mg / mL using Amicon® filter (10 kd cut off) in PBS 7.4 (200 pL final volume). This was followed by addition of 4 equivalences of linker-payload (2.7pL of 19 from 10 mM stock) in DMSO and the same Eppendorf® tube was incubated at 24 °C for 3 hours. Complete conjugation was confirmed by deconvolution of intact mass-spec revealing mostly DAR (drug to antibody ratio) of 2. The excess linker-payload was removed by using Cytiva™, G-25 (PD-10 column, 0.5mL load and 1 mL elute) column. The eluted fraction was reconcentrated to 5 mg / mL using Ami con" filter (lOkd cut off) in desired stability buffer (5 mg / mL final concentration ADC).

[0210] Step 2: The Cys-engineered IgGl antibody conjugated with 19 (0.85 mg in 200 pL) was buffer exchanged into 100 mM Tris 10 mM EDTA in PBS 7.4 (pH 8.2) in 1.5 ml Eppendorf® tube. Six equivalences of reductant 2 (4 pL from 10 mM stock solution in water) and volume was adjusted using the same buffer to final concentration of 5 mg / mL. The Eppendorf® tube was incubated on thermomixer at 37-40 °C for 2 hours under shaking conditions (800-1000 rpm). Complete reduction of S375C cysteine was confirmed by Intact mass-spec analysis demonstrating mass loss of 2 cysteines.

[0211] After reduction, the excess of the reductant was removed using Cytiva™, G-25 (PD-10 column, 0.5 mL load and 1 mL elute) column. The eluted fraction was reconcentrated to 5 mg / mL using Amicon® filter (10 kd cut off) in PBS 7.4 (200 pL final volume). This was followed by addition of 4 equivalences of linker-payload (2.7pL of 28 from 10 mM stock) in DMSO and the same Eppendorf® tube was incubated at 24 °C for 3 hours. Complete conjugation was confirmed by deconvolution of intact mass-spec revealing mostly DAR (drug to antibody ratio) of 4. The excess linker-payload was removed by using Cytiva™, G-25 (PD-10 column, 0.5 mL load and 1 mL elute) column. The eluted fraction was reconcentrated to 5 mg / mL using Amicon® filter (10 kd cut off) in desired stability buffer (5 mg / mL final concentration ADC).4) Protocol for IdeS digestion as a proof of selective reduction at S375C or S168C

[0212] To 10 pg of ADC in PBS 7.4 was added 10 units of FabRICATOR® (IdeS) and reaction was maintained at 37 °C for 1 hour in 1.5mL Eppendorf® tube, this was followed by excess TCEP addition (20 equivalences from 0.5 M stock solution). Reaction was shaken on thermomixer for additional 60 minutes and the sample were analyzed by intact mass-spec under reverse phase LC- MS conditions.5) Cys-engineered antibody preparation a) Cell culture and stable pool transfection

[0213] Suspension adapted Chinese hamster ovary' (CHO) K-l (CHO-K1) cells with a glutamine synthetase gene knockout (Horizon Discovery Group pic, Cambridge, UK) were grown in chemically defined (CD) CHO Medium (Gibco®, Thermo Fisher Scientific, Waltham, MA USA) supplemented with HT (5 mM sodium hypoxanthine and 0.8 mM thymidine; Gibco®) and 8 mM glutamine (Gibco®) at 37 °C and 5% CO2. Stable transfections were carried out using an electroporation method (Maxcyte®, Rockville, MD USA) according to the vendor’s protocol.Cells were co-transfected with 59 pg SPB HC DNA. 59 pg SPB LC DNA and 7 pg Transposase DNA (Hera Biolabs, Lexington, KY US A) in Maxey te® EP Buffer (SPB: Succinimidyl-[4-(psoralen-8-yloxy)]-butyrate). Transfected cells were then allowed to recover in growth media for 2 days and then were split into methionine sulfoximine (Sigma- Aldrich®) containing selection media. Cells were passaged in selection media until they were fully recovered. Once stable pools were established, fed-batch production was initiated. During production, cells from the established stable pools were seeded into Dynamis™ Media (Gibco®) supplemented with HT (Gibco®) and CuSO4 (Sigma-Aldrich®). Cells were then grown in continuous culture for 12 days with bi-weekly feeds of glucose and Efficient Feed™ C AGT™ (Gibco®). The cell cultures were then harvested, and the supernatant clarified using a 0.2 pm filter. Subsequent filtered supernatant was then purified. b) Transient transfection and antibody production

[0214] Suspension adapted CHO-Expi cells (EpiCHO™; Thermo Fisher Scientific) were grown in CHO-Expi Growth medium (Thermo Fisher Scientific) at 37 °C and 5% CO2. Transient transfections were carried out according to CHO-Expi Max -Titer protocol (Thermo Fisher Scientific). In this method, cells were co-transfected with 10 pg plasmid expression vector encoding the heavy chain and 10 pg plasmid expression vector encoding the light chain using Expifectamine™ (Thermo Fisher Scientific) transfection reagent. Transient transfected cells were then grown in continuous culture for 7 days with feeds on day 1 and day 5 and temperature-shift on day 1 to 32 °C. On day 7, supernatant was harvested and clarified using a 0.2 pm filter. Subsequent supernatant was then purified. c) Cys-engineered antibody purification

[0215] Cys-engineered antibodies were first purified from supernatant by affinity chromatography using MabSelect™ PrismA™ protein A resin (Cytiva™). The pH of the eluted antibodies were then adjusted to 3.5 using 1 M acetic acid and incubated at room temperature for 1 hour. The pH was then adjusted to 7.0 using 1 M Tris base followed by anion exchange chromatography purification using Capto™ Q resin (Cytiva™). The antibodies were then polished using either cation exchange chromatography (Capto™ SP ImpRes, Cytiva™) or size exclusion chromatography (Superdex™ 200, Cytiva®). The purified antibodies w ere then buffer exchanged into the final formulation buffer, 20 mM sodium acetate, 9% sucrose, pH 5.5. The purity of the final samples was tested using a BEH200 UPLC-SEC analytical column (Waters Corporation), reduced and nonreduced CE-SDS (PerkinElmer), and reverse-phase-HPLC.Endotoxin level was also quantified using Endosafe® nexgen-MCS™ (Charles River Laboratories International, Inc., Fairfield, NJ USA). For Cys-engineered Cys-mutants, thepurified antibodies went through an additional Cysteine deblocking process using a two-step reduction / re-oxidation method before buffer exchanging into the formulation buffer. Briefly, the polished antibodies were first buffer exchanged into IX PBS at pH 7.4 followed by a reduction step using dithiothreitol (DTT. Thermo Fisher Scientific) at a 30: 1 DTT: antibody molar ratio for 2 hours at room temperature. The samples were then extensively dialyzed against IX PBS to remove the residual DTT. The reduced antibodies were then re-oxidized using dehydroascorbic acid (DHAA, Sigma- Aldrich®) at a 20: 1 DHAA: antibody molar ratio overnight at 4°C followed by dialysis into the formulation buffer. The level of remaining Cysteinylation / Glutathionylation was then determined using LC-MS.5) Details of computational methods for the investigation of pKa, solvent accessibility, and reagent accessibility

[0216] Accessibility of phosphine reductants towards each individual disulfide sites was evaluated computationally through an in-house developed workflow that integrates small molecule docking and Rosetta FastRelax (Rosetta Commons Org ). Docking was performed with Molecular Operating Environment (MOE) (Version 2022.02, Chemical Computing Group (CCG)) using Assisted Model Building with Energy Refinement (AMBER) 14-EHT forcefield. Full-length IgG model was prepared from Protein Data Bank (PDB) structure ID: 1HZH. Missing densities and S375C / S168C with cysteine cap were modeled in the IgG structure by MOE. Individual phosphines w ere first relaxed under AMBER14-EHT, and then docked into a 5 A- radius sphere centered on each disulfide cysteine pair. Docking calculations were performed with triangle matcher for initial placement, London dG for placement scoring, and a maximum of 1000 poses were kept for refinement. Refinement was performed using the Induced Fit option, with tethering weight for side chains set at 10, cutoff distance set at 6 A, gradient / iterations set at 0.01 / 500, and Generalized-Bom Volume Integral / Weighted Surface area (GBVI / WSA) dG used as scoring method. After refinement, a maximum of 50 poses in total were kept. Each docking calculation was performed 5 times for every phosphine respectively against S375C-cap, S168C- cap, H-L site, and H-H site.

[0217] Rosetta (Version 3. 13, Rosetta Commons Org.) FastRelax was used to optimize each final pose. Parameter files of the reductants and cysteine capped S375C or S168C for Rosetta inputs were prepared by the molfile_to_params.py script, using mol2 output from MOE after minimization under Merck Molecular Force Field 94x (MMFF94x) forcefield (Halgren, J. Comput. Chem. 17: 490-519 (1996)). Rotamer libraries of the phosphines were generated as PDB rotamers from Low ModeMD conformational search outputs by MOE. LowModeMD w as set upunder MMFF94x forcefield, with rejection limit at 100. Root Mean Square (RMS) gradient at 0.005, interaction / MM interaction limit at 10000 / 500, Root Mean Square Deviation (RMSD) limit at 0.25 A, energy window at 7 kcal / mol, and conformation limit at 10000. Constraints were placed on the relative geometry of the docked phosphine molecule and target disulfide. A preliminary quantum mechanical (QM) geometry optimization was performed on a model phosphine-disulfide pair (trimethylphosphine and dimethyl disulfide), using b31yp / 6-31 g(d) calculations. The converged geometry suggested an optimal distance between the reactive phosphorus and sulfur to be 4.5 A, and every S-P-C angle larger than 90 degrees. Following this observation, the Rosetta constraints were set up as follows: a 4.5 A harmonic constraint was placed between P and the ‘'reactive” S, and four flat harmonic constraints were placed respectively on the three S-P-C angles and one S-S-P angle at 90-180 degrees. Two constraints files were used with each S being the “reactive” sulfur. FastRelax calculation was set up by “- relax: fast -relax: constrain_relax_to_start_coords -exl -ex2 -use_input_sc -flip_HNQ -no_optH false -relax: cartesian -score: set_weights cart_bonded 0.5 pro_close 0 -constraints :cst_fa_weight 100.0 -nstruct 2”, with an additional movemap and resfile that included the phosphine, disulfide, and amino acids w ithin 10 A of the disulfide for sidechain repack, backbone flexibility, and phosphine-antibody “jump” flexibility. For each phosphine-disulfide simulation, all final relaxed structural outputs were used to extract the summed “fa dun” scores for antibody amino acids included in the movemap, averaged across the entire structure ensemble, and subsequently averaged across all 5 repeat runs to represent the accessibility7of the phosphine towards the disulfide site. Attempts to numerically represent phosphine-disulfide accessibility7were also made after the docking stage (i.e., before FastRelax). P-S distances were measured for each docked pose, averaged across all poses, and subsequently averaged across all 5 repeats to represent the accessibility.

[0218] The whole docking / relax / data-processing process has been w ritten as a series of automated bash and python scripts workflow and is ready to be routinely run for computationally characterizing new phosphine designs.

[0219] pKas of disulfide sites of interest w ere calculated by Propka3 (GNU Lesser General Public License v2.1), using structures where the cysteines of interest w ere transformed into the reduced form and diversified by FastRelax (Cartesian minimization, nearby 10 A amino acids included). Average and standard deviation values were reported. To evaluate disulfide solvent accessibility, molecular dynamics (MD) simulation was performed by OpenMM (OpenMM Org., Version 7.7.0) with graphics processing unit (GPU) acceleration (NVIDIA® P100, Compute Unified Device Architecture (CUD A) version 10.2.89; Nvidia Corporation, Santa Clara, CAUSA). The “pdbfixer” tool was used to standardize the S375C and S168C mutated IgG model, and then calculation was set up with AMBER 14, gbn2 implicit solvent, 2 nm non-bonded cutoff, and H bond constraints. 100 ns simulation was performed at 300 K using Langevin integrator after energy minimization and equilibration. Snapshots were taken every 0.5 ns during production run, stored for trajectory analysis, and then had the engineered cysteines transformed to the capped version by Rosetta FastDesign (Cartesian minimization, 10 A nearby amino acids included). Solvent accessibility of the sulfur atoms of each disulfide of interest was calculated by PyMOL (Version 2.3.1, Schrodinger LLC, New York, NY, USA).EXAMPLE 1Experimental procedures and1H-NMR spectrum for synthesized phosphine (22)

[0220] Procedure: To a stirred solution of dicyclohexylphosphane (5 g, 25.2 mmol) in toluene (50 mL) were added 4-bromobenzoic acid (6.08 g, 30.3 mmol), 1,1'- bis(diisopropylphosphino)ferrocene (0.211 g, 0.504 mmol), sodium tert-butoxide (3.64 g, 37.8 mmol) and palladium(ii) acetate (0. 113 g, 0.504 mmol) under nitrogen atmosphere at 25 °C. The reaction was stirred at 80 °C for 12 hours. The reaction mixture was concentrated under reduced pressure, crude compound was diluted with 10% methanol (MeOH) in dichloromethane (DCM) (200 mL), filtered, washed with 10% MeOH in DCM (100 mL) and filtrate was concentrated under reduced pressure. Crude compound was purified by column chromatography using Biotage® 12 g silica gel cartridge and compound eluted with 10% MeOH in DCM. Pure fractions were combined, concentrated under reduced pressure and crude compound was purified by prep- HPLC purification. Pure fractions were combined and lyophilized to afford 22 (64 mg, 0.195 mmol, Liquid chromatography-mass spectrometry (LC-MS): 96.83%).Prep, high performance liquid chromatography (HPLC) conditions:

[0221] MOBILE PHASE - 10 mM Ammonium Bicarbonate in H2O: methyl cyanide (MeCN); COLUMN - X-Bridge C18 Pack (19X250), 5p Flow-19.0 mL / minutes;

[0222] GRADIENT METHOD - 0 / 10.2 / 10,8 / 55, 13 / 73.6, 13.05 / 100.16 / 100,16.05 / 10, 19 / 101

[0223] LCMS (ESI) calcd. for (C19H27O2P) [M-HJ- 317. 17. found 317.25, Rt = 4.52 minutes;

[0224] 'H NMR (400 MHz, DMSO-de) 5 = 7.88 (d. J = 7.8 Hz. 2H), 7.49 (dd, J = 6.4, 7.9 Hz. 2H), 1.95 (br t, J = 11.4 Hz, 2H), 1.84 - 1.67 (m, 4H), 1.67 - 1.49 (m, 6H), 1.37 - 1.15 (m, 4H), 1.12 - 0.96 (m, 4H), 0.95 - 0.77 (m, 2H).EXAMPLE 2Experimental procedures and1H-NMR spectrum for synthesized phosphine (23)

[0225] Step 1 : To a stirred solution of 4-iodobenzenesulfonic acid (1.576 g, 5.55 mmol), Sodium tert-butoxide (0.582 g, 6.05 mmol), l.l'-Bis(diisopropylphosphino)ferrocene (DIPPF) (65.15 mg, 0.121 mmol) in toluene (15 mL) was added Palladium (II) acetate (0.226 g. 1.009 mmol) at room temperature and stirring continued for 60 minutes. Then to this reaction mixture Dicyclohexylphosphane (1.0 g, 5.04 mmol) was added at room temperature. The reaction mixture was stirred in sealed tube at 80 °C for 16 hours. (Notes: The above reaction was set up in glove box and heating maintained in fume hood under nitrogen atmosphere). Reaction mixture was concentrated under reduced pressure. Crude compound was diluted with 10% MeOH in DCM (20 mL) and stirred at room temperature for 10 minutes. Solid was filtered and dried under vacuum. Crude compound was purified by prep-HPLC purification. Pure fractions were combined and lyophilized. The obtained compound was again triturated with n-pentane (3 x 2 mL) and dried under reduced pressure. Again, obtained compound re-purified by prep-HPLC purification. Pure fractions were combined and lyophilized to afford 23 (85 mg, 0.216 mmol, LCMS: 90.98%).Prep-HPLC Conditions I:

[0226] MOBILE PHASE - 10 mM Ammonium Bicarbonate in H2O: MeCN COLUMN - X- Bndge C18 Pack (25X250), 5p Flow - 19.0 mL / minutes. GRADIENT METHOD: 0 / 10, 1.5 / 10, 8 / 63, 13.44 / 63, 13.5 / 100, 16.5 / 100, 16.51 / 10, 21 / 10 Prep-HPLC Conditions II:

[0227] COLUMN: Kromacil Pack 150*25 mm 5p Mobile Phase A: 5 mM ABC in water Mobile Phase B: (85: 15) acetonitrile (ACN):MeOH METHOD (T / %B) 0 / 55.7.5 / 55 7.55 / 100 FLOW RATE: 19 mL / minutes APMS-007

[0228] LCMS (ESI) calcd. for (C18H27O3PS) [M+H]+355. 14. found 355.23, Rt = 4.52 min. 'H NMR (400 MHz, DMSO-de) 5 = 7.58 (d, J= 11.6 Hz, 2H), 7.40 (d, J= 7.2 Hz, 2H), 1.98-1.91 (m, 2H), 1.80-1.68 (m, 4H), 1.62-1.51 (m, 6H), 1.31-0.99 (m, 10H).EXAMPLE 3Experimental procedures and1H-NMR spectrum for synthesized phosphine (24)

[0229] Step-1 : To a stirred solution of bis(4-fluorophenyl)phosphine oxide (2 g, 8.40 mmol) in degassed Tetrahydrofuran (THF) (20 mL) was added 1 M Diisobutylaluminium hydride (DIBAL-H) (16.79 mL, 25.2 mmol) drop-wise at room temperature. The reaction mixture was stirred at 25 °C for 1 hour. The reaction mixture was diluted with Methyl tert-butyl ether (MTBE) (60 mL) at 0 °C. Then to this mixture degassed 2N NaOH (40 mL) solution was added slowly followed by brine (50 mL). Organic layer was separated under nitrogen atmosphere, dried over Na2SO4, filtered, and concentrated under reduced pressure with nitrogen atmosphere to afford compound 2 (1.8 g. Crude).

[0230] Step 2 : To a stirred solution of 4-iodobenzenesulfonic acid (2.301 g, 8.10 mmol) in toluene (30 mL) were added Sodium tertiary butoxide (0.934 g. 9.72 mmol), 1,1'- Bis(diisopropylphosphino)ferrocene (0.102 g. 0.243 mmol), Palladium(ii) acetate (0.045 g, 0.203 mmol) at room temperature in glove box. Reaction mixture was stirred at 25°C for 1 hour. Then to this reaction mixture was added compound 2 (5069340-0383-001, 1.8 g, Crude) and stirred at 80 °C for 12 hours. The reaction mixture was diluted with diethyl ether (50 mL) and solid was filtered. Solid was dissolved in 10% MeOH in DCM (60 mL) and ACN (20 mL). Solid was filtered and filtered was concentrated under reduced pressured to afford crude compound (1.9 g). The crude compound was purified by reverse phase column chromatography and compound eluted with 40% acetonitrile (CAN) in water. Pure fractions were combined and concentrated under reduced pressure. Obtained compound was again purified by reveres phase prep HPLC purification. Pure fractions were combined and lyophilized to afford 24 (155 mg. 0.397 mmol. LCMS: 97.40%).

[0231] LCMS (ESI) calcd. for (C18H13F2O3PS) [M-H]- 377.03, found 377.11, Rt = 4.19 min.

[0232] 'H NMR (400 MHz, DMSO-de) 5 (ppm) = 7.61 (d, J= 7.2 Hz, 2H), 7.29-7.24 (m, 8H),7.18 (t, J= 8.2 Hz, 2H).EXAMPLE 4Experimental procedures and1H-NMR spectrum for synthesized phosphine (25)

[0233] Step 1 : To a stirred solution of bis(4-methoxyphenyl)phosphine oxide (2 g, 7.63 mmol) in degassed THF (20 mL) was added IM DIBAL-H (22.88 mL, 22.88 mmol) drop-wise at room temperature. The reaction mixture was stirred at 25 °C for 1 hour. The reaction mixture was diluted with degassed MTBE (60 mL) and cooled to 0 °C. Then to this reaction mixture degassed 2N NaOH (40 mL) solution was added slowly followed by brine (50 mL). Organic layer was separated under nitrogen atmosphere, dried over Na2SO4, filtered, and concentrated under nitrogen flushing to afford compound 2 (1.8 g, Crude)

[0234] Step 2: To a stirred solution of 4-bromobenzenesulfonic acid (1.733 g, 7.31 mmol) in toluene (30 mL) was added Sodium tert-butoxide (0.843 g, 8.77 mmol), 1,T- Bis(diisopropylphosphino)ferrocene (0.092 g, 0.219 mmol), Palladium(ii) acetate (0.041 g, 0.183 mmol) in glove box. Reaction mixture was stirred at 25 °C for 1 hour. Then to this reaction mixture compound 2 (5069340-0317-002, 1.8 g, Crude) was added at room temperature. The reaction mixture was stirred at 80 °C for 12 hour. Reaction mixture was diluted with ether (50 mL) and filtered the reaction mixture. Solid was dissolved in 10% MeOH in DCM (60 mL) and ACN (20 mL). Again, solid was filtered and filtrate was concentrated under reduced pressure. Crude compound was purified by reverse phase column chromatography and compound eluted with 40% ACN in water. Pure fraction was concentrated under reduced pressure to afford 25 (155 mg, 0.351 mmol, LCMS: 91.27%).

[0235] LCMS (ESI) calcd. for (C20H19O5PS) [M+H]+403.07, found 403. 16, Rt = 3.48 mm.1H NMR (400 MHz, DMSO-de) 5 = 7.56 (dd, J= 1.4, 8.3 Hz, 2H), 7.22 - 7.15 (m, 4H), 7.15 - 7.08 (m. 2H), 6.98 (dd, J= 0.8, 8.8 Hz, 4H), 3.76 (s, 6H).EXAMPLE 5Experimental procedures and1H-NMR spectrum for synthesized phosphine (26)

[0236] Step 1: To a stirred solution of cyclohexyl magnesium bromide (11.0 g. 58.8 mmol) in THF (100 mL) was added ethyl phenylphosphinate (10.0 g, 58.8 mmol) dropwise slowly at 0 °C. The reaction mixture was stirred at 25 °C for 2 hours. Reaction mixture was quenched with NT Cl solution (200 mL), water (300 mL) and extracted with DCM (600 mL). Combined organic layer was washed with brine (300 mL), dried over anhydrous sodium sulphate, filtered, and concentrated under reduced pressure. Crude compound was purified by Biotage using silica gel (230-400 mesh) column and compound eluted with 50% ethyl acetate in pet ether. Pure fractions were combined and concentrated under reduced pressure to afford 3 (4.00 g, 17 mmol, LCMS: 91.72%).

[0237] LCMS (ESI) calcd. for (C12H17OP) [M+H]+209.10, found 209.17, Rt = 0.81 mm.

[0238] Step 2: To a stirred solution of 1.5M DIBAL-H (19.2 mL, 28.8 mmol) (1.5 M IN Toluene) in degassed THF (20 mL) was added 3 (2.00 g, 9.60 mmol) dropwise at 65 °C. Then the reaction mixture was stirred at 65 °C for 1 hour. Reaction mixture was diluted with degassed MTBE (40 mL) and cooled to 0 °C. Then degassed 2N NaOH (40 mL) solution and brine solution (10 mL) slowly were added to the reaction mixture at room temperature. Combined organic layer under nitrogen atmosphere was dried over Na2SC>4, filtered, and concentrated under nitrogen flushing to afford 4 (1.50 g, 0.5 mmol, LCMS: 6.29%).

[0239] LCMS (ESI) calcd. for (C12H17P) [M+H]+193. 11. found 193.04, Rt = 1.35 min.

[0240] Step 3 : To a stirred solution of 4-iodobenzoic acid (1.64 g, 6.63 mmol) in toluene (15 rnL) was added sodium 2-methylpropan-2-olate (1.00 mL, 9.36 mmol), palladium diacetate (52.6 mg, 234 pmol), cyclopentyl(diisopropyl)phosphane; iron (97.9 mg, 234 pmol) at room temperature in glove box. The reaction mixture was stirred at 25 °C for 1 hour. Then 4 (1.50 g, 7.80 mmol) was added to the reaction mixture at room temperature. The reaction mixture was stirred at 80 °C for 16 h. Reaction mixture was diluted with ether (50 mL) and filtered. Obtained solid was dissolved in 10% MeOH in DCM (60 mL) &ACN (20 mL). Then reaction mixture was filtered and filtrate was concentrated under reduced pressured. Crude compound was purified by Reverse phase column chromatography and compound eluted with 40% ACN in water.

[0241] Pure fractions were combined, concentrated under reduced pressure and obtained compound was further re-purified by Prep HPLC purification under cooling condition (conditions: ATLANTIES T3: MOBILE PHASE - lOmM Ammonium Bicarbonate IN H2O: ACN: MEOH (70:30) COLUMN - ATLANTIES T3 (19X250) mm 5u Flow-15 mL / minutes GRADIENT METHOD - 0 / 38,2 / 38,12 / 65,15 / 65,15.05 / 100,17 / 100,17.05 / 38,20 / 38). Pure fractions were combined, concentrated under reduced pressure, and lyophilized to afford 26 (35.0 mg, 0.11 mmol, LCMS: 95.90%).

[0242] LCMS (ESI) calcd. for (C19H21O2P) [M-H]’ 311.13, found 311.16, Rt = 4.36 min.

[0243] 'H NMR (400 MHz, DMSO-de): 5 (ppm) 7.84 (d, J= 7.2 Hz, 2H), 6.96-7.76 (m, 8H), 2.28-2.40 (m, 1H), 1.48-1.80 (m, 5H), 1.00-1.40 (m, 5H).EXAMPLE 6Experimental procedures and1H-NMR spectrum for synthesized phosphine (27)

[0244] Step 1: To a stirred solution of 4-iodobenzenesulfonic acid (1.88 g, 6.63 mmol) in toluene (15 rnL) was added sodium 2-methylpropan-2-olate (1.00 rnL, 9.36 mmol), palladium diacetate (52.6 mg, 234 pmol), cyclopentyl(diisopropyl)phosphane; iron (97.9 mg, 234 pmol) atroom temperature in glove box. The reaction mixture was stirred at 25 °C for 1 hour. Then 4 (1.50 g, 7.80 mmol) was added to the reaction mixture at room temperature. The reaction mixture was stirred at 80°C for 16 hours. Reaction mixture was diluted with ether (30 mL) and filtered. Obtained solid was dissolved in 10% MeOH in DCM (60 mL) & ACN (20 mL). Reaction mixture was filtered, and filtrate was concentrated under reduced pressure. Crude compound was purified by Prep HPLC purification under cooling condition.

[0245] MOBILE PHASE - 1 OmM Ammonium Bicarbonate IN H2O: MeCN COLUMN - X- Bridge, C18 (19X250) mm, 5p Flow-15.0 mL / minutes.

[0246] GRADIENT METHOD - 0 / 35,4 / 35,12 / 70,12.05 / 100,16 / 100,16.05 / 35,20 / 35). Pure fractions were combined, concentrated under reduced pressure, and lyophilized to afford 27 (88.0 mg, 0.23 mmol, LCMS: 92.17%).

[0247] LCMS (ESI) calcd. for (C18H21O3PS) [M+H]+349.09, found 349. 16, Rt = 4.38 min.

[0248] 'H NMR (400 MHz, DMSO-de): 5 (ppm) 7.57 (d, J= 7.6 Hz, 2H), 7.28-7.50 (m, 7H), 7.07 (s. 4H), 2.25-2.38 (m. 1H), 1.50-1.80 (m. 5H), 1.00-1.40 (m. 6H).EXAMPLE 7Experimental procedures and!H-NMR spectrum for synthesized phosphine (29)

[0249] Step 1 : To a stirred solution of 4-(diphenylphosphaneyl)-N,N-dimethylaniline (2.0 g, 6.55 mmol) in THF (20 mL) was added BH3.THF (7.86 mL, 7.86 mmol) at -5 °C. The reaction mixture was stirred at 0 °C for 2 hours. Reaction mixture was concentrated under reduced pressure. The residue was dissolved in DCM (20 mL) and added Mel (2.457 mL. 39.3 mmol) at room temperature. The reaction mixture was stirred at reflux temperature for 5 days.!H NMR showed 50% formation of compound. Then methyl iodine (Mel) (1 mL) was added to the reaction mixture at room temperature. The reaction mixture was stirred at reflux for 2 days. Reaction mixture was filtered, and filtrate was concentrated under reduced pressure to afford 2 (1.1 g, 2.33 mmol. LCMS: 98.25%).

[0250] LCMS (ESI) calcd. for (C21H27BINP) [M-(I+BH3)]+320. 10, found 320.29, Rt = 0.78 minutes.

[0251] Step 2: To a stirred solution of 2 (900 mg. 1.947 mmol) in methanol (2 mL) was added dicyclohexylamine (0.388 mL, 1.947 mmol) at room temperature. The reaction mixture was at reflux for 16 h. Reaction mixture was cooled to room temperature to get precipitate. Precipitate was washed with EtOAc: pet ether (1 : 1) (3 x 5 mL) and water (3 x 5 mL) and again washed with EtOAc: Pet ether (1 : 1) (3 x 5 mL) to obtained compound. Obtained compound was dried under vacuum to afford pure desired product (220 mg). But LCMS showed 82.38% (18.85+63.53). Compound was again washed with water (3 x 5 mL) and EtOAc: pet ether (1: 1) (3 x 5 mL). Obtained compound was dried under vacuum to afford 29 (200 mg. 0.431 mmol, LCMS: 96.50%).

[0252] LCMS (ESI) calcd. for (C21H23INP) [M-I]+320.06, found 320.17, Rt = 0.26 mm & 1.64 min.

[0253] 'H NMR (400 MHz, DMSO-de): 5 (ppm) 7.98 (d, J= 8.4 Hz, 2H), 7.37-7.53 (m, 8H), 7.21-7.35 (m, 4H), 3.59 (s. 9H).EXAMPLE 8Experimental procedures and1H-NMR spectrum for synthesized phosphine (30)

[0254] Step 1 : To a stirred reagent of (4-fluorophenyl) magnesium bromide (64.6 mL, 64.6 mmol) was added a solution of ethyl phenylphosphinate (5.0 g, 29.4 mmol) in THF (20 mL) at - 78 °C. The reaction mixture was stirred at 25 °C for 2 hours. Reaction mixture was cooled to 0 °C. Reaction mixture was quenched with sat. NH4CI solution (60 mL) and extracted with EtOAc(3 x 25 mL). Combined organic layer was washed with brine (15 mL), dried over anhydrous sodium sulphate, filtered, and concentrated under reduced pressure. Crude compound was purified by Biotage® using 80 g silica (100-200 mesh) cartridge and compound eluted with 80% EtOAc in pet ether. Pure fractions were combined and concentrated under reduced pressure to afford 3 (2.5 g. 11.35 mmol. LCMS: 76.68%).LCMS (ESI) calcd. for (C12H10FOP) [M+H]+221.05, found 221.19, Rt = 0.77 min.

[0255] Step 2 : To a stirred solution of DIBAL-H (18. 17 mL, 27.3 mmol) in toluene was added a solution of 3 (2.0 g, 9.08 mmol) in THF (5 mL) at 60 °C (preheated oil bath). The reaction mixture was stirred at 60 °C for 1 hour. Reaction mixture was cooled to -78 °C and quenched with 2N NaOH solution (20 mL). To Reaction mixture was added MTBE (20 mL) and stirred for 10 minutes. Organic layer w as separated with syringe and dried over sodium sulphate. Again, organic layer was transferred into another R.B with syringe and concentrated with nitrogen flashing under argon atmosphere to afford 4 (2.0 g, 9.80 mmol), which was used directly for next step without purification.

[0256] Step 3 : To a stirred solution of 4-iodobenzenesulfonic acid (2.087 g, 7.35 mmol) in toluene (8 mL) were added sodium tert-butoxide (1.035 g, 10.77 mmol), 1,1'- bis(diisopropylphosphino)ferrocene (0.205 g, 0.490 mmol) and palladium (II) acetate (0.088 g, 0.392 mmol) at room temperature in a sealed tube in glove box. The reaction mixture was stirred at 25 °C for 1 hour. Then a solution of 4 (2.0 g, 9.80 mmol) in toluene (2 mL) was added to the reaction mixture at room temperature. The reaction mixture was stirred at 80 °C for 16 hours in fume hood. Reaction mixture was cooled to room temperature. Reaction mixture was diluted with pet ether and filtered. Obtained compound was again diluted with 10% MeOH in DCM (15 mL). again filtered and filtrate w as concentrated under reduced pressure. Crude compound was purified by reverse phase column (Biotage®) using 80 g Cl 8 column and compound eluted with 50% ACN in w ater. Pure fractions w ere combined, concentrated under reduced pressure, and obtained compound was further re-purified by Prep HPLC purification:

[0257] MOBILE PHASE: 20 mM Ammonium Bicarbonate in H2O: MeCN Column - X-Bridge packed C18 (25X250) mm, 5p Flow-15.0 mL / minute.

[0258] GRADIENT METHOD: -0 / 37,2 / 37,15.0 / 70,15.05 / 100,18 / 100,18.05 / 37,20 / 37). Pure fractions were combined, concentrated under reduced pressure, and lyophilized to afford 30 (147 mg, 0.396 mmol. LCMS: 97.06%).

[0259] LCMS (ESI) calcd. for (CisHuFOsPS) [M+H]+361.04, found 361.10, Rt = 4.24 mm.

[0260] ‘H NMR (400 MHz, DMSO-d6): 5 (ppm) 7.58-7.65 (m, 2H), 7.37-7.46 (m, 3H), 7.15- 7.34 (m, 8H), 7.07 (s, 4H).EXAMPLE 9Identification and optimization of chemoselective reductants through phosphine screening

[0261] In our investigation, we focused on assessing the selective reduction ability of 18 commercially available phosphine-based compounds (Fig. 2A). We specifically examined their effectiveness in reducing the engineered cysteine site at S375C in the Fc region or S168C in the Fab region (Fig. 2B) and conjugated vedotin (19) as a model linker-payload. These sites were chosen because they represent distinct local environments within the structure of the antibody. The ADCs produced using each reductant were subjected to LCMS analysis to calculate the average drug-to-antibody ratio (DAR). In addition, we employed IdeS digestion followed by subunit LCMS analysis to determine the precise site of conjugation (Fig. 2D and Fig. 2E). The selective reduction capabilities of the reductants were assessed using a non-reducing CE-SDS (Table 2, Table 3, and Table 4), which provided valuable insights into their ability to selectively reduce the engineered cysteine sites of interest. Reductants that exhibited less than 90% main peak area in the non-reducing CE-SDS profile were classified as non-selective, indicating their potential to reduce inter-chain disulfide bonds (Table 2). Conversely, reductants with greater than 90% main peak area were classified as selective, reflecting their abil i ty to selectively reduce the engineered cysteine sites. TCEP. a known phosphine reductant with the ability to non- selectively reduce all disulfide bonds, was included as a reference.

[0262] Our observations revealed that reductants 2, 3, 4, 5, 6, 7, and 11 (Table 2, Fig. 2C and Fig. 2E) displayed remarkable selectivity7in reducing the S375C engineered cysteine site. These reductants demonstrated intact ADCs of over 90% (Table 2) and achieved a DAR within the range of 1.7 to 2.0 (Table 2). Interestingly, these same reductants also exhibited selective reduction capabilities towards the S168C site (Fig. 2A), which is in a distinct domain of the antibody. The classification of selective versus non-selective remained consistent across both sites (Table 2), highlighting the broad applicability of these reductants in selectively reducing engineered cysteine sites throughout various regions of the antibody structure. However, we observed that reductants 1, 8, 9, 10, 12, and 13, although capable of avoiding the reduction of inter-chain disulfide bonds, did not achieve complete conjugation (Table 2). Despite intact ADCs of greater than 90% according to the CE-SDS analysis, a DAR range of 0.3 to 1.6 indicated incomplete reduction (Table 2). We were, however, able to achieve complete conjugation with reductants 1 and 9 by increasing the stoichiometry of the reductant (Fig. 21A - Fig. 21B). Our observations revealed a consistent behavior of these reductants across a range of antibodies with diverse sequences and targets which provides evidence that these findings can be applied broadly to all human IgGl isotype monoclonal antibodies. Some of the reductants we studied have been previously investigated for the preparation of site-specific ADCs. Notably. Coumans et al.(Bioconjug. Chem. 31 : 2136-2146 (2020)) reported that reductants 1, 2, and 4 exhibited complete reduction of the antibody, while reductant 8 selectively reduced the engineered cysteine inside the Fab cavity7at the HC-41C site. Conversely, Liao et al., op. cit., observed that reductant 1 did not reduce inter-chain disulfide bonds. These differences in findings can be attributed tovariations in reaction conditions and reductant stoichiometry. In our study, we have demonstrated that the selectivity of reductants 2 and 4 can be controlled by optimizing the reductant stoichiometry. pH, time, and temperature. Hence, to gain a deeper understanding of the observed behavior and investigate the underlying mechanisms, it is crucial to compare the activities of these reductant under identical and scalable reaction conditions. To ensure a rigorous evaluation, we optimized and standardized the conjugation conditions (Fig. 2C) for all the reductant disclosed herein as detailed in the methods section. We conducted a series of in-depth computational and reduction rate kinetics experiments to gain insights into the governing principles that dictate the reduction trends.EXAMPLE 10Computational modeling of phosphine selectivity

[0263] Because the data herein shows that a broad selectivity profile may be achieved with phosphine-based reductants, multiple computational characterizations were performed to investigate mechanisms underlying this selectivity and establish predictive descriptor for future development of selective reductants. We first analyzed whether there were intrinsic differences between the capped engineered disulfides and the native interchain and intrachain disulfides. pKa was previously reported to be correlative to disulfide stability (Vollmar et al., Bioconjug. Chem. 28: 2538-2548 (2017)). We calculated the pKa values (Olsson et al., J. Chem.l Theory Comput. 7: 525-537 (2011)) of the cysteines at sites S375C, S168C, and the interchain disulfides respectively at the Fab interface (referred to as the HL site) and the heavy chain hinge interface (referred to as the HH site) (Fig. 3B). No significant difference was found in the predicted pKa values (Fig. 3A). suggesting that the disulfides of interest have similar potential for reduction. We then calculated solvent accessibility of the disulfides using molecular dynamic simulation. Sites with more solvent exposure are intrinsically easier to be reduced, but the engineered cysteines displayed lower solvent accessibility to the sulfur atoms than the HL disulfide. These findings suggest that selective reduction ability of the reductants depend on factors other than pKa and solvent accessibility differences of cysteines (Fig. 3B).

[0264] The data suggests that selective reduction may be controlled by site-specific phosphine accessibility, reactivity, or a combination of both. Initially, a preliminary docking-based reagent accessibility analysis was performed on S375C and HL sites, without involving HH disulfides located on the highly flexible IgG hinge. Rigid-backbone docking was performed to sample favorable poses of each phosphine near the disulfides. The distance between the reactive phosphorous and sulfur atoms for the poses was summarized to represent phosphine accessibility'(Fig. 20A - Fig. 20C). and the difference of accessibility for each compound against HL and S375C sites was reported. A clear differentiation of the experimentally observed non-selective and selective reductants was observed for HL vs. S375C (Fig. 20A - Fig. 20C), suggesting that site-specific reagent accessibility is a main contributor to the selective reduction. To further validate this and improve the modeling workflow, the docked pose for each phosphine was refined in a 2ndstep calculation by Rosetta FastRelax protocol (Tyka et al., J. Molec. Biol. 405: 607-618 (2011)), with large flexibility' introduced to nearby amino acids backbone and side chains during the constrained relaxation, with large flexibility introduced to nearby amino acids backbone and side chains during the constrained relaxation. The Rosetta fa dun score term (Dunbrack’s statistical potential of protein side chain conformational favorability) (Alford et al., J. Chem. Theory Comput. 13: 3031-3048 (2017)) from relaxed structures was extracted for nearby amino acids to evaluate the energy strain introduced by approaching phosphine to the protein environment, as another way to represent reagent accessibility (Fig. 20A - Fig. 20C).

[0265] We incorporated high backbone flexibility in the 2ndstep calculations to capture the behavior of HH disulfide accurately. With this updated workflow, statistical difference in accessibility scores between HH and capped S375C cysteine disulfide bonds was observed, which enabled us to successfully distinguish selective and non-selective reductants (Fig. 3C). The distinction between selective and non-selective reductants was not apparent when the HH backbone was treated as fixed, underscoring the importance of incorporating backbone flexibility in our analysis. The accessibility' descriptors from the docking and constrained relaxation steps were correlative with each other, suggesting that shared physical properties were represented. Encouraged by the successful modeling of selective S375C reduction behaviors of the reductants, we applied the same docking-relaxation accessibility modeling protocol to predict the selectivity on S168C, an alternative engineered cysteine site on Fab domain. We observed a clear differentiation of selective and non-selective groups by summed accessibility' differences between HH / HL and capped S168C thiols (Fig. 3C). Notably, TCEP. commonly used non- selective disulfide reductant, exhibited the lowest relative accessibility between engineered cysteines and interchain disulfides (Fig. 3C). Overall, the above computational campaigns provided multiple evidence that suggest reagent accessibility plays a major role in TPP selective reduction, and computational descriptors of accessibility' can be a robust predictor for selective reduction of engineered cysteines against IgG interchain disulfides. We developed a modeling workflow that accurately predicted the chemoselective behavior of reductants for two engineered cysteine sites located in distinct regions of antibody. This workflow provides an in-silico tool for assessing the selectivity' of a wide range of engineered cysteines.EXAMPLE 11Studies on the intrinsic chemical reactivity! of phosphine reductants and reduction rate kinetic analysis

[0266] Despite initially favorable computed accessibility scores, reductants 8, 10, 12, and 13 exhibited incomplete conjugations (Table 2), suggesting a partial reduction of the engineered cysteine sites. To investigate the reasons behind this inefficiency,31P NMR experiments were designed to measure the reduction kinetics of these reductants (Fig. 3E and Fig. 3F).31P NMR kinetics measurements for reductants

[0267] 100 mM reductant stock solutions were prepared by d-DMSO. 100 mM L-cysteine methylester dihydrochloride (L-CME) solution was prepared by d-DMSO / PBS buffer mixture (v / v = 50 / 50). To an NMR tube was added 300 pL of reductant solution (100 mM, 1 Eq), followed by 300 pL L-cystine dimethyl ester dihydrochloride solution (100 mM. 1 Eq). The NMR tube was quickly flipped upside down to evenly mix the solutions and NMR experiments (Bruker, 400 MHz, No. of scan = 32, delay = 5s) were started immediately after the addition of L-CME solution. The reaction kinetics were monitored at 5-minute intervals for an hour. The baseline signal at To was acquired by reductant stock solution diluted by d-DMSO / PBS buffer to 50 mM using the same instrumental setting. The reductant concentration at each time point was determined by normalizing its area under the curve wi th respect to the signal at To, and the reaction rate constant was fitted by second order rate kinetics using the following equation:

[0268] Where [A]o is the initial concentration of phosphine reductants, [A] is the concentration at the given time point t. k is the rate constant of the reduction and is calculated through linear regression analysis.

[0269] Figs. 6 to 16 show the ' 'P NMR kinetics measurements for reductants 1, 2, 3, 4, 5, 6, 8, 9, 11, and 15

[0270] In these experiments, L-cysteine methylester was used as a model disulfide substrate and monitored the real-time depletion of the reductants using31P NMR. The cleavage of disulfide bonds by the reductants involves a rate-limiting SN2-type nucleophilic attack from the phosphine onto one of the sulfur atoms, displacing a sulfhydryl group and forming a cationic thiophosphonium ion (Henderson. Jr. and Buckler, J. Amer. Chem. Soc. 82: 5794-5800 (1960): Humphrey and Hawkins, Analyt. Chem. 36: 1812-1814 (1964); Mthembu et al., ChemBioChem21 : 1947-1954 (2020)). Subsequent hydrolysis rapidly releases the second thiol fragment and generates phosphine oxide (Fig. 3E). Reduction rates were calculated using a second-order rate kinetics model, enabling clear differentiation among the different types of reductants (Figs. 6 to 16 and Table 5).

[0271] Interestingly, non-selective reductants, primarily those with biphenyl substitution (14, 15, 16, 17, and 18), exhibited higher reduction rates (> 0.1 M-ls-1), likely due to lowered steric hindrance caused by flexible alky l backbone and inductive effects. Conversely, the presence of electron-withdrawing sulfonyl or carboxylic substitutions decreased the reduction rate (4, 2, 5, 8 > 3 > 1). Reductants with sulfonyl groups at the para-position exhibited slightly faster reaction kinetics compared to ortho-substituted phosphines (e.g., 8). On the other hand, electron-donating groups at the para-position (e.g., 6 and 11) accelerated the reduction rate. Due to limited solubility in the DMA and buffer mixture, the rate kinetics for reductants 10, 12, and 13 using31P NMR could not be determined. Hence, LCMS was employed as an alternative method to study the reduction rate kinetics of these reductants. Reductants 10 and 12 exhibited rare and unexpected stability of the thiophosphonium species (Overman and O'Connor, J. Amer. Chem. Soc. 98: 771-775 (1976)). Through above mechanistic observation, chemoselectivity was discovered to be greatly influenced by electronic effects of specific functional groups (e.g., ortho substituted aldehydes) leading to stable thiophosphonium complexes that are not susceptible to hydrolysis (Fig. 3E), providing an explanation for the lower DAR observed with these reductants.

[0272] Combining previous modeling efforts, site accessibility differences and chemical reactivity appear to be critical factors for achieving site-specific and complete uncapping. Forinstance, although reductant 15 exhibited similar reduction kinetics compared to other selective reductants (Fig. 3E and Fig. 3F), its limited distinguishable site accessibility difference between interchain disulfide and S375C / S168C compromised its site-selectivity. Conversely, reductant 1, despite having the highest site accessibility difference, failed to achieve effective uncapping of the antibodies due to its relatively slow reduction kinetics (Fig. 3F and Fig. 6). These observations offer valuable guidance for the design and screening of next generation chemoselective reductants for uncapping any engineered cysteine site.

[0273] By using accessibility descriptor models and reactivity of 1stgeneration reductants, we designed novel phosphine-based reductants (Fig. 3D) that may have greater controlled reactivity. For example, attaching quaternary amine group (reductant 29) on to the para position and combining strongly / moderately electron withdrawing groups (reductants 24 and 30) such as sulfate and fluorine’s resulted in higher accessibility and greater decapping selectivity against S375C and S168C (Table 6). Interestingly, reductants 24 and 29 possess greater controlled reactivity even at higher equivalence (Table 6) of phosphine (24 equivalence) without compromising the selective decapping efficiency making these reductants ideal for large-scale synthesis of site-specific ADCs.EXAMPLE 12Dual-drug ADC synthesis using orthogonally reactive phosphine-based reagents.

[0274] Combining drugs with different toxicity profiles and modes of action is a common practice in cancer therapy, and as a result, several ADCs are being used in combination with clinically approved drugs to enhance treatment efficacy (Fanale et al., J. Clin. Oncol. 32: 3137- 3143 (2014): Younes et al., The Lancet Oncol. 14: 1348-1356 (2013)). However, a more appealing approach to deliver multiple drugs simultaneously is through the functionalization of an antibody with payloads that have distinct mechanisms of action (MoA). This innovative approach opens new possibilities to improve the clinical success of ADCs, particularly in cases where targeting low or heterogeneous levels of tumor-associated antigens and drug resistance is challenging. Numerous research groups have explored the synthesis and evaluation of dual-drug ADCs (Kumar et al., Bioorg. Med. Chem. Letts. 28: 3617-3621 (2018); Levengood et al., Angew Chem. Int. Ed. Engl. 56: 733-737 (2017); Nilchan et al., Antibody Thera. 2: 71-78 (2019);Yamazaki et al., Nat. Comm. 12: 3528 (2021)), revealing their potential for enhanced therapeutic efficacy compared to the co-administration of single-drug ADCs. These approaches typically involve modifying antibodies to incorporate orthogonal conjugation handles (Y amazaki et al., op. cit.) for attaching different drugs or designing multifunctional linkers capable of accommodating distinct drugs (Levengood et al., op. cit.). However, these methods lack modularity and often necessitate extensive antibody engineering or the development of complex multifunctional linkers to achieve efficient synthesis of dual-payload ADCs.

[0275] Through the screening experiments, reductant 8 was discovered to exhibit a distinct reactivity pattern towards S168C at higher stoichiometry, while avoiding S375C engineered cysteine site. This observation led the design of a method for preparing dual payload ADCs in a stepwise manner in which the S168C site is selectively reduced using reductant 8, followed by the reduction of the S375C (Fig. 4A) site using reductant 2. Consistent with our findings, reductant 8 exhibited selective reduction of the S168C site at higher stoichiometry (See U.S. Pat. No. 10.814,009). Fig. 4A demonstrates the controlled conjugation of payloads 19 and 21 to the two different engineered sites, guided by the selective reduction profiles of representative reductants. The successful conjugation of the dual payload with a DAR 4 was confirmed throughIdeS digestion followed by LCMS subunit analysis (Fig. 4C). Importantly, this demonstrated that dual conjugation at two different antibody sites can be achieved without reducing the interchain disulfides. The resulting DAR 4, dual-payload ADC predominantly existed as a monomeric species, with 98% abundance (Fig. 4D). This innovative and efficient approach provides a streamlined and modular method for the rapid screening and evaluation of multiple drug combinations. It enables the ability to systematically assess and identify promising dual-payload ADC platforms that exhibit comparable toxicities while leveraging the distinct mechanisms of action of the individual drugs.EXAMPLE 13

[0276] The computational modeling of phosphine selectivity based on Example 10 is presented below.Step 1: Structure preparation for molecular docking.

[0277] Full-length IgG model was prepared from PDB structure ID: 1HZH. Missing densities and engineered cysteines (S375C or S168C) with cysteine cap were modeled in the IgG structure by MOE (Version 2022.02, CCG), by sidechain replacement and minimization.Step 2: Docking of phosphine of interest to engineered and interchain disulfides.

[0278] Docking was performed with MOE (Version 2022.02, CCG) using Amberl4-EHT forcefield. Individual phosphines were first relaxed under Amber 14-EHT, and then docked into a 5 A-radius sphere centered on each disulfide cysteine pair. Docking calculations were performed with triangle matcher for initial placement, London dG for placement scoring, and a maximum of 1000 poses were kept for refinement. Refinement was performed using the Induced Fit option, with tethering weight for side chains set at 10, cutoff distance set at 6 A, gradient / iterations set at 0.01 / 500, and GBVI / WSA dG used as scoring method. After refinement, a maximum of 50 poses in total were kept. Each docking calculation w as performed 5 times for every phosphine respectively against engineered cysteine with cap (S375C-cap or S 168C-cap. for example) and native interchain disulfides betw een antibody heavy and light chains (H-L site), and between heavy and heavy chains (H-H site).Step 3: Optimization of docked poses from step 2

[0279] Rosetta (Version 3.13, RosettaCommons) FastRelax was used to optimize each final pose. Parameter files of the phosphine compounds and cysteine capped S375C or S168C for Rosetta inputs were prepared by the molfile_to_params.py script, using mol2 output from MOEafter minimization under MMFF94x forcefield. Rotamer libraries of the phosphines were generated as PDB rotamers from LowModeMD conformational search outputs by MOE. LowModeMD was set up under MMFF94x forcefield, with rejection limit at 100, RMS gradient at 0.005, iteraction / MM iteraction limit at 10000 / 500, RMSD limit at 0.25 A. energy window at 7 kcal / mol. and conformation limit at 10000. Constraints were placed on the relative geometry of the docked phosphine molecule and target disulfide. Rosetta constraints were set up as follows: a 4.5 A harmonic constraint w as placed between P and the “reactive” S, and four flat harmonic constraints were placed respectively on the three S-P-C angles and one S-S-P angle at 90-180 degrees. Two constraints files were used with each S being the “reactive” sulfur. FastRelax calculation was set up by “-relax:fast -relax:constrain_relax_to_start_coords -exl -ex2 - use_input_sc -flip_HNQ -no_optH false -relax: cartesian -score: set weights cart_bonded 0.5 pro_close 0 -constraints :cst_fa_weight 100.0 -nstruct 2”, with an additional movemap and resfile that included the phosphine, disulfide, and amino acids within 10 A of the disulfide for sidechain repack, backbone flexibility, and phosphine-antibody “jump” flexibility. Two output structures w ere produced for each saved dock pose.Step 4: Calculation of phosphine-disulfide accessibility difference between engineered and interchain disulfides, to represent selectivity of the modeled phosphine towards the engineered cysteine.

[0280] All final relaxed structural outputs from step 3 w ere used to extract the summed “fa_dun” scores for antibody amino acids included in the Rosetta movemap, averaged across the entire structure ensemble, and subsequently averaged across all 5 repeat runs to represent the accessibility of the phosphine towards the disulfide site. To numerically represent phosphinedisulfide accessibility after the docking stage (i.e.. before FastRelax). P-S distances were measured for each docked pose, averaged across all poses, and subsequently averaged across all 5 repeats to represent the accessibility. Phosphine selectivity towards the engineered cysteine site was represented as difference between the accessibility scores to interchain disulfides (H-H and H-L) and engineered cysteine.

[0281] The disclosed subject matter is not to be limited in scope by the specific embodiments and examples described herein. Indeed, various modifications of the disclosure in addition to those described will become apparent to those skilled in the art from the foregoing description and accompanying figures. Such modifications are intended to fall within the scope of the appended claims.

[0282] All references (e.g.. publications or patents or patent applications) cited herein are incorporated herein by reference in their entirety and for all purposes to the same extent as if each individual reference (e.g., publication or patent or patent application) was specifically and individually indicated to be incorporated by reference in its entirety for all purposes. Other embodiments are within the following claims.

Claims

WHAT IS CLAIMED IS:

1. A method for selectively reducing a capped engineered cysteine residue over interchain disulfide bonds of a Cys-engineered antibody to produce a reduced, uncapped Cys- engineered antibody, the method comprising contacting the Cys-engineered antibody with a reductant selected from the group consisting of: 4-(diphenylphosphaneyl)benzenesulfonic acid (2), 4,4'-(phenylphosphanediyl)dibenzenesulfonic acid (3),3-(diphenylphosphaneyl)benzenesulfonic acid (4),4-(diphenylphosphaneyl)benzoic acid (5), 4-(diphenylphosphaneyl)-N,N-dimethylaniline (6), (R)-2-((diphenylphosphaneyl)methyl)pyrrolidine (7), 2-(diphenylphosphaneyl)benzenesulfonic acid (8), triphenylphosphine (9), 2-(diphenylphosphaneyl)benzaldehyde (10), tris(4-methoxyphenyl)phosphane (11), 2-(di-p-tolylphosphaneyl)benzaldehyde (12), 2-(dicyclohexylphosphaneyl)benzenesulfonic acid (13), 4-(bis(4-fluorophenyl)phosphaneyl)benzenesulfonic acid (24), tris(4-fluorophenyl)phosphane (28), 4-(diphenylphosphaneyl)-N,N,N-trimethylbenzenaminium (29), 4-((4-fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30), and salts thereof. for a time sufficient to reduce the capped engineered cysteine of the Cys-engineered antibody to produce the reduced, uncapped Cys-engineered antibody.

2. The method of claim 1. wherein the Cys-engineered antibody comprises one or more engineered cysteine residues, wherein each engineered cysteine residue is located at a position within the Cys-engineered antibody selected from the group consisting of heavy chain positions 152, 153, 171, 172, 173, 347, 362, 375, 380, 382, 393, 430, and 440; wherein the position number is according to EU numbering; and / or the group consisting of heavy chain positions 40, 41, and 89; wherein the position number is according to Kabat numbering.

3. The method of claim 1, wherein the Cys-engineered antibody comprises one or more engineered cysteine residues, wherein each engineered cysteine residue is located at aposition within the Cys-engineered antibody selected from the group consisting of light chain positions 1 5 and 168; wherein the position number is according to EU numbering; and / or the group consisting of light chain positions 40 and 41; wherein the position number is according to Kabat numbering.

4. The method of claim 1 , wherein the Cys-engineered antibody comprises (a) a first engineered cysteine residue located at a position selected from the group consisting of heavy chain positions 152, 153, 171, 172, 173, 347, 362. 375, 380, 382, 393, 430, and 440; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue located at a position selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering.

5. The method of claim 1. wherein the Cys-engineered antibody comprises (a) a first engineered cysteine residue at heavy chain position 375; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue at light chain position 168; wherein the position number is according to EU numbering.

6. The method of claim 1. wherein the reduced, uncapped Cys-engineered antibody is separated from the reductant and the reduced, uncapped Cys-engineered antibody is incubated with a payload or linker-payload comprising a thiol reactive group for a time sufficient to produce a conjugate comprising the Cys-engineered antibody conjugated to the payload or linkerpayload.

7. A method for producing a composition of antibody conjugates, the method comprising:(a) providing a composition of Cys-engineered antibodies, wherein each antibody comprises (i) two heavy chains and two light chains, (ii) one or more capped engineered cysteine residues, and (iii) interchain disulfide bonds between the hinge region of the two heavy chains and between the hinge region of the heavy chain and the light chain;(b) mixing the composition with an excess of reductant selected from the group consisting of: 4-(diphenylphosphaneyl)benzenesulfonic acid (2), 4,4'-(phenylphosphanediyl)dibenzenesulfonic acid (3),3-(diphenylphosphaneyl)benzenesulfonic acid (4),4-(diphenylphosphaneyl)benzoic acid (5),4-(diphenylphosphaneyl)-N.N-dimethylaniline (6), (R)-2-((diphenylphosphaneyl)methyl)pyrrolidine (7), 2-(diphenylphosphaneyl)benzenesulfonic acid (8), triphenylphosphine (9), 2-(dipheny Iphosphaney l)benzaldehy de ( 10). tris(4-methoxyphenyl)phosphane (11),2-(di-p-tolylphosphaneyl)benzaldehyde (12),2-(dicyclohexylphosphaneyl)benzenesulfonic acid (13), 4-(bis(4-fluorophenyl)phosphaneyl)benzenesulfonic acid (24), tris(4-fluorophenyl)phosphane (28), 4-(diphenylphosphaneyl)-N,N,N-trimethylbenzenaminium (29), 4-((4-fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30), and salts thereof. to provide a reaction mixture;(c) incubating the reaction mixture for a time sufficient to selectively reduce the capped engineered cysteines of the Cys-engineered antibodies over the interchain disulfide bonds of the Cys-engineered antibodies to produce reduced, uncapped Cys-engineered antibodies;(d) separating the reduced, uncapped Cys-engineered antibodies from the reductant to provide a composition of separated, reduced, uncapped Cys-engineered antibodies; and(e) incubating the composition of separated, reduced, uncapped Cys- engineered antibodies with an excess of payload comprising a thiol reactive group for a time sufficient to produce the composition of antibody conjugates.

8. The method of claim 7. wherein the Cys-engineered antibodies comprise one or more engineered cysteine residues, wherein each engineered cysteine residue is located at a position within the Cys-engineered antibody selected from the group consisting of heavy chain positions 152, 153, 171, 172, 173, 347, 362, 375, 380, 382, 393, 430, and 440; wherein the position number is according to EU numbering; and / or the group consisting of heavy chain positions 40, 41, and 89; wherein the position number is according to Kabat numbering.

9. The method of claim 7, wherein the Cys-engineered antibodies comprise one or more engineered cysteine residues, wherein each engineered cysteine residue is located at aposition within the Cys-engineered antibody selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering; and / or the group consisting of light chain positions 40 and 41; wherein the position number is according to Kabat numbering.

10. The method of claim 7, wherein the Cys-engineered antibodies comprise (a) a first engineered cysteine residue located at a position selected from the group consisting of heavy chain positions 152, 153, 171, 172, 173, 347, 362. 375, 380, 382, 393, 430, and 440; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue located at a position selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering.

11. The method of claim 1. wherein the Cys-engineered antibodies comprise (a) a first engineered cysteine residue at heavy chain position 375; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue at light chain position 168; wherein the position number is according to EU numbering.

12. The method of claim 7. wherein the excess of reductant comprises 6 equivalences of reductant per engineered cysteine.

13. The method of claim 7, wherein the excess of payload comprises 4 equivalences of payload per engineered cysteine.

14. The method of claim 7, wherein greater than 90% of the Cys-engineered antibodies in the composition of antibody conjugates are intact.

15. The method of claim 7. wherein the composition of antibody conjugates has a drug-to-antibody ratio (DAR) between 1.7 and 2.0.

16. A method for producing a composition of Cys-engineered antibodies having a tetrameric structure in which the engineered cysteine residues thereof are selectively reduced over interchain disulfide bonds of the Cys-engineered antibodies, the method comprising(a) providing a composition of Cys-engineered antibodies comprising capped engineered cysteines and a reductant selected from the group consisting of: 4- (diphenylphosphaneyl)benzenesulfonic acid (2), 4,4'-(phenylphosphanediyl)dibenzenesulfonic acid (3),3-(diphenylphosphaneyl)benzenesulfonic acid (4),4-(diphenylphosphaneyl)benzoic acid (5), 4-(diphenylphosphaneyl)-N,N-dimethylaniline (6), (R)-2-((diphenylphosphaneyl)methyl)pyrrolidine (7), 2-(diphenylphosphaneyl)benzenesulfonic acid (8), triphenylphosphine (9), 2-(diphenylphosphaneyl)benzaldehyde (10), tris(4-methoxyphenyl)phosphane (11), 2-(di-p-tolylphosphaneyl)benzaldehyde (12), 2-(dicyclohexylphosphaneyl)benzenesulfonic acid (13). 4-(bis(4-fluorophenyl)phosphaneyl)benzenesulfonic acid (24), tris(4-fluorophenyl)phosphane (28), 4-(diphenylphosphaneyl)-N,N,N-trimethylbenzenaminium (29), 4-((4-fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30), and salts thereof;(b) mixing the composition with the reductant to provide a reaction mixture; and(c) incubating the reaction mixture for a time sufficient to reduce the capped engineered cysteines of the Cys-engineered antibody to produce the reduced, uncapped Cys- engineered antibody.

17. The method of claim 16, wherein the Cys-engineered antibodies comprise one or more engineered cysteine residues, wherein each engineered cysteine residue is located at a position within the Cys-engineered antibody selected from the group consisting of heavy chain positions 152, 153, 171, 172, 173, 347, 362, 375, 380, 382, 393, 430, and 440; wherein the position number is according to EU numbering; and / or the group consisting of heavy chain positions 40, 41, and 89; wherein the position number is according to Kabat numbering.

18. The method of claim 16, wherein the Cys-engineered antibodies comprise one or more engineered cysteine residues, wherein each engineered cysteine residue is located at aposition within the Cys-engineered antibody selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering; and / or the group consisting of light chain positions 40 and 41; wherein the position number is according to Kabat numbering.

19. The method of claim 16, wherein the Cys-engineered antibodies comprise (a) a first engineered cysteine residue located at a position selected from the group consisting of heavy chain positions 152, 153, 171, 172, 173, 347, 362. 375, 380, 382, 393, 430, and 440; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue located at a position selected from the group consisting of light chain positions 165 and 168; wherein the position number is according to EU numbering.

20. The method of claim 16, wherein the Cys-engineered antibodies comprise (a) a first engineered cysteine residue at heavy chain position 375; wherein the position number is according to EU numbering; and (b) a second engineered cysteine residue at light chain position 1 8; wherein the position number is according to EU numbering.

21. The method of claim 16, wherein greater than 90% of the reduced, uncapped Cys- engineered antibodies in the composition are intact.

22. A method for producing a composition of antibody conjugates conjugated to a first and a second payload, the method comprising:(a) providing a mixture comprising Cys-engineered antibodies having a heavy chain and a light chain, a first capped engineered cysteine residue at position 1 8 of the light chain, a second capped engineered cysteine residue at position 375 of the heavy chain, and interchain disulfide bonds, wherein the position numbers are according to EU numbering;(b) adding to the mixture an excess of first reductant 2- (diphenylphosphaneyl)benzenesulfonic acid (8) or salt thereof to produce a first reducing mixture, and incubating the first reducing mixture for a time sufficient to selectively reduce the first capped engineered cysteines at position 168 of the light chains without reducing the interchain disulfide bonds of the Cys-engineered antibodies;(c) separating the first reduced, uncapped Cys-engineered antibodies from the first reductant to provide a first composition of Cys-engineered antibodies;(d) adding an excess of a first payload comprising a thiol reactive group to the first composition to produce a first conjugation mixture and incubating the first conjugation mixture for a time sufficient to produce Cys-engineered antibodies conjugated to the first payload;(e) separating the Cys-engineered antibodies conjugated to the first payload from the first conjugation mixture to provide a composition of Cys-engineered antibodies conjugated to the first pay load;(f) adding to the composition of Cys-engineered antibodies conjugated to the first payload an excess of a second reductant 4-(diphenylphosphaneyl)benzenesulfonic acid (2) or salt thereof to produce a second reducing mixture, and incubating the second reducing mixture for a time sufficient to selectively reduce the second capped engineered cysteines at position 375 of the heavy chains without reducing the interchain disulfide bonds of the Cys-engineered antibodies to produce second reduced, uncapped Cys-engineered antibodies;(g) separating the second reduced, uncapped Cys-engineered antibodies from the second reductant to provide a second composition of Cys-engineered antibodies;(h) adding an excess of a second payload comprising a thiol reactive group to the second composition to produce a second conjugation mixture and incubating the second conjugation mixture for a time sufficient to produce Cys-engineered antibodies conjugated to the second payload; and(i) separating the Cys-engineered antibodies conjugated to the second payload from the second conjugation mixture to provide a composition of Cys-engineered antibodies conjugated to the first payload and the second payload.

23. The method of claim 22, wherein the excess of the first reductant comprises about 40 equivalences of reductant per capped engineered cysteine.

24. The method of claim 22 or 23, wherein the excess of the second reductant comprises about 6 equivalences of reductant per capped engineered cysteine.

25. The method of claim 22, wherein the excess of pay load comprises 4 equivalences of payload per reduced engineered cysteine.

26. A dual-payload antibody conjugate comprising an antibody having a light chain comprising a cysteine residue at position 168 conjugated to a first payload and a heavy chaincomprising a cysteine residue at position 375 conjugated to a second payload, wherein the position numbering is according to Eu numbering and wherein the first payload and the second payload are not the same.

27. An antibody conjugate comprising an antibody having a light chain and a heavy chain, wherein the light chain comprises an engineered cysteine at position 168 conjugated to a first payload and the heavy chain comprises an engineered cysteine residue at position 375 conjugated to a second pay load, and wherein the position numbering is according to Eu numbering and wherein the first payload and the second payload are not the same.

28. A reductant selected from the group consisting of: 4-(bis(4- fluorophenyljphosphaneyljbenzenesulfonic acid (24), 4-(diphenylphosphaneyl)-N,N,N-trimethylbenzenaminium (29), 4-((4- fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30), and salts thereof.

29. A reductant selected from the group consisting of: 4- (dicyclohexylphosphaneyl)benzoic acid (22), 4-(dicyclohexylphosphaneyl)benzenesulfonic acid (23), 4-(bis(4-methoxyphenyl)phosphaneyl)benzenesulfonic acid (25), 4- (cyclohexyl(phenyl)phosphaneyl)benzoic acid (26), 4- (cyclohexyl(phenyl)phosphaneyl)benzenesulfonic acid (27), and salts thereof.

30. A method for selectively reducing capped engineered cysteine residues of an antibody comprising incubating the antibody with an excess of a phosphine-based reductant for a time sufficient to reduce the capped engineered cysteine residues, wherein the phosphine-based reductant comprises 4-(bis(4-fluorophenyl)phosphaneyl)benzenesulfonic acid (24), 4- (diphenylphosphaneyl)-N,N,N-trimethylbenzenaminium (29), 4-((4- fluorophenyl)(phenyl)phosphaneyl)benzenesulfonic acid (30), or salts thereof.

31. A method for reducing capped engineered cysteine residues of an antibody comprising incubating the antibody with an excess of a phosphine-based reductant for a time sufficient to reduce the capped engineered cysteine residues, wherein the phosphine-based reductant is selected from the group consisting of: 4-(dicyclohexylphosphaneyl)benzoic acid (22), 4-(dicyclohexylphosphaneyl)benzenesulfonic acid (23), 4-(bis(4-methoxyphenyl)phosphaneyl)benzenesulfonic acid (25), 4-(cyclohexyl(phenyl)phosphaneyl)benzoic acid (26), 4-(cyclohexyl(phenyl)phosphaneyl)benzenesulfonic acid (27), and salts thereof.

Images