Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 290 (12), 7535-62

A Novel Antibody Engineering Strategy for Making Monovalent Bispecific Heterodimeric IgG Antibodies by Electrostatic Steering Mechanism

Affiliations

A Novel Antibody Engineering Strategy for Making Monovalent Bispecific Heterodimeric IgG Antibodies by Electrostatic Steering Mechanism

Zhi Liu et al. J Biol Chem.

Abstract

Producing pure and well behaved bispecific antibodies (bsAbs) on a large scale for preclinical and clinical testing is a challenging task. Here, we describe a new strategy for making monovalent bispecific heterodimeric IgG antibodies in mammalian cells. We applied an electrostatic steering mechanism to engineer antibody light chain-heavy chain (LC-HC) interface residues in such a way that each LC strongly favors its cognate HC when two different HCs and two different LCs are co-expressed in the same cell to assemble a functional bispecific antibody. We produced heterodimeric IgGs from transiently and stably transfected mammalian cells. The engineered heterodimeric IgG molecules maintain the overall IgG structure with correct LC-HC pairings, bind to two different antigens with comparable affinity when compared with their parental antibodies, and retain the functionality of parental antibodies in biological assays. In addition, the bispecific heterodimeric IgG derived from anti-HER2 and anti-EGF receptor (EGFR) antibody was shown to induce a higher level of receptor internalization than the combination of two parental antibodies. Mouse xenograft BxPC-3, Panc-1, and Calu-3 human tumor models showed that the heterodimeric IgGs strongly inhibited tumor growth. The described approach can be used to generate tools from two pre-existent antibodies and explore the potential of bispecific antibodies. The asymmetrically engineered Fc variants for antibody-dependent cellular cytotoxicity enhancement could be embedded in monovalent bispecific heterodimeric IgG to make best-in-class therapeutic antibodies.

Keywords: Antibody; Antibody Engineering; Cancer Therapy; Fc-γ Receptor; Pancreatic Cancer.

Figures

FIGURE 1.
FIGURE 1.
Residues at VH-VL and CH1-CL interfaces for substitutions are buried, conserved, and spatially close. Anti-HER2 trastuzumab crystal structure 1N8Z from Protein Data Base is illustrated as an example. Structure 1N8Z pdb was loaded into Molecular Operating Environment (Chemical Computing Group, Montreal, Canada). Amber 10:EHT force field was set. Missing atoms were repaired, and charges were applied to the termini as appropriate at pH 7.0 automatically by the structure preparation function. Ribbon rendering was selected with a different color for each of the four domains in the following manner: VH, magenta; VL, orange; CH1, blue; Cκ1, green. Selected residues were rendered as ball-and-stick, and the Site View Function isolated a region within 5 Å of all selected residues. Kabat numbering was used for the variable region residues, whereas Eu numbering was used for the constant region residues. Distances were measured between each αC of the indicated (circled) residue pair by the distance tool. A, side view of VH-VL. The selected residues Gly-44, Gln-39, and Gln-105 in VH; Gln-100, Gln-38, and Ala-43 in VL are buried in the hydrophobic core of VH-VL. CDR loops are positioned at the top. B, VH-VL interface. Circled pairs are substituted with charged residues to drive the electrostatic steering effect. C, side view of CH1-Cκ. The selected residues Pro-171, Ser-183, and Ala-141 in CH1 and Ser-162, Ser-176, and Phe-116 in Cκ are buried in the hydrophobic core of CH1-Cκ. D, CH1-Cκ interface. Residues Pro-171, Ser-183, and Ala-141 in CH1 and Ser-162, Ser-176, and Phe-116 in Cκ are in proximity, respectively. Lys-147 in CH1 and Gln-124/Ser-131/Thr-180 in Cκ are located in the middle of hydrophobic core (as Ser-183 in CH1 and Ser-176 in Cκ) but are not shown for view simplicity. Circled pairs are substituted with charged residues to drive the electrostatic steering mechanism. E, configuration of monovalent bispecific hetero-IgG antibody variants using the electrostatic steering approach. Two pairs of charged residues, KK-DD, in the variable regions binding to antigen A combined with one pair of charged residues, D-K, in CH1-CL drives the LC1 to pair with its cognate HC1.Similarly, two pairs of charged residues, DD-KK, in the variable regions binding to antigen B combined with one pair of charged residues, K-D, in CH1-CL drives the LC2 to pair with its cognate HC2. The charged residues for heterodimerization in the CH3 domains are also indicated. F, configuration of monovalent bispecific hetero-IgG antibody variants using the electrostatic steering approach. Two pairs of charged residues, KK-DD, in the variable regions binding to antigen A combined with two pairs of charged residues, DD-KK, in CH1-CL drives the LC1 to pair with its cognate HC1.Similarly, two pairs of charged residues, DD-KK, in the variable regions binding to antigen B combined with two pairs of charged residues, KK-DD, in CH1-CL drives the LC2 to pair with its cognate HC2. The charged residues for heterodimerization in the CH3 domains are also indicated. More configurations can be found in patent application WO2014081955. The symbol “⊕” represents positively charged residues and “⊝” represents negatively charged residues.
FIGURE 2.
FIGURE 2.
Proof-of-concept studies to validate the feasibility of hetero-IgG format. Anti-HER2 × EGFR hetero-IgG1 variants in which an Fn3 tag was attached to the N terminus of anti-EGFR HC2 and an Fn3-FLAG-His6 tag was attached to the C terminus of anti-EGFR LC2 were made and tested by multiple assays. A, dual antigen-binding plate ELISA. The sequence variations of anti-HER2 × EGFR hetero-IgG variants 1C02, 1C04, 2A05, 2B05, and 5D03 are indicated in Table 2. B, Western blotting of hetero-IgG variants. M, molecular weight standard. Lane 1, transfectants from wild type anti-HER2 IgG1 HC and LC. Lane 2, transfectants from anti-HER2 HC1 and LC1 together with anti-EGFR HC2 and LC2. Lane 3, anti-HER2 × EGFR hetero-IgG variant 1C02. Lane 4, anti-HER2 × EGFR hetero-IgG variant 1C04. Lane 5, anti-HER2 × EGFR hetero-IgG variant 2A05. Lane 6, anti-HER2 × EGFR hetero-IgG variant 2B05. Lane 7, anti-HER2 × EGFR hetero-IgG variant 5D03. C, purified anti-HER2 × EGFR hetero-IgG variants 1C02, 1C04, 2A05, 2B05, and 5D03 were subjected to electrophoresis in 8–16% SDS-PAGE under nonreducing conditions and then stained with Coomassie Blue. The molecular mass standards are labeled at the left of the bands in kDa. D, purified anti-HER2 × EGFR hetero-IgG variants 1C02, 1C04, 2A05, 2B05, and 5D03 were subjected to electrophoresis in 8–16% SDS-PAGE under reducing conditions and then stained with Coomassie Blue. The molecular mass standards are labeled at the left of the bands in kDa. E, ADCC killing assay of anti-HER2 × EGFR hetero-IgG1 variants 2B05 and 5D03 and irrelevant human IgG1 using NCI-N87 target cells and NK effector cells having FcγR IIIA 158F/F genotype. F, percent inhibition of pEGFR by anti-EGFR E7.6.3 IgG1 alone. G, percent inhibition of pEGFR by the combination of anti-HER2 humAb4D5–8 IgG1 and anti-EGFR E7.6.3 IgG1. H, percent inhibition of pEGFR by the anti-HER2 × EGFR hetero-IgG1 variant 2B05. I, percent inhibition of pEGFR by the anti-HER2 × EGFR hetero-IgG1 variant 5D03. J, percent inhibition of basal pHER3 by anti-HER2 humAb4D5–8 IgG1 alone. K, percent inhibition of basal pHER3 by the combination of anti-HER2 humAb4D5–8 IgG1 and anti-EGFR E7.6.3 IgG1. L, percent inhibition of basal pHER3 by the anti-HER2 × EGFR hetero-IgG1 variant 2B05. M, percent inhibition of basal pHER3 by the anti-HER2 × EGFR hetero-IgG1 variant 5D03.
FIGURE 3.
FIGURE 3.
Hetero-IgG1 variant 2B05 in the presence of Fn3 and Fn3-FLAG-His6 tags has the predicted mass and correct LC-HC pairings by mass spectrometry analysis. A, intact hetero-IgG1 after deglycosylation by PNGase F. B, anti-HER2 LC1. C, anti-EGFR LC2 tagged with Fn3-FLAG-His6 at the C terminus. D, anti-HER2 HC1. E, anti-EGFR HC2 tagged with Fn3 at the N terminus. F, partial reduction of hetero-IgG1 by 2-fold of TCEP. Different components include two HCs (HC1 + HC2), ¾ Ab (LC1 + HC1 + Fn3-HC2), ¾ Ab (HC1 + Fn3_HC2 + LC2_Fn3-FLAG-His6). Residual intact hetero-IgG is also present. G, ½ Ab consisting of anti-HER2 LC1 and HC1. H, ½ Ab consisting of anti-EGFR Fn3_HC2 and LC2_Fn3-FLAG-His6. I, depiction of products of hetero-IgG1 variant 2B05 that were partially reduced by TCEP. The main peak G0F at 74,062.65 daltons for anti-HER2 HC1+LC1 and at 96,888.25 daltons for anti-EGFR Fn3_HC2 + LC2_Fn3_FLAG_His6 has monosaccharide composition of (GlcNAc)2(Man)3(GlcNAc)2Fuc; G1F has monosaccharide composition of Gal(GlcNAc)2(Man)3(GlcNAc)2Fuc; G2F has monosaccharide composition of (Gal)2(GlcNAc)2(Man)3(GlcNAc)2Fuc; the peaks at 73,859.08 daltons and 96,670.75 daltons miss a GlcNAc and have the monosaccharide composition of GlcNAc(Man)3(GlcNAc)2Fuc. Intensity (ion counts/s) represents the y axis in A–H.
FIGURE 4.
FIGURE 4.
Chain drop-out transient transfections to assess the LC-HC pairing tolerances for hetero-IgG1 variants in the absence of any tags. 2936E cells were transfected with either two or four different plasmid DNAs. Six days post-transfection, crude supernatant was loaded in 8–16% Tris-glycine SDS-polyacrylamide gel and subjected to electrophoresis under nonreducing conditions and Western blotting. The sequence variations of anti-HER2 × EGFR hetero-IgG1 variants 2B05 and 5D03 are indicated in Table 2. Variants V15, V20, V21, V22, V23, and V25 are indicated in Table 3. LC1 and HC1 are derived from anti-HER2 trastuzumab, and LC2 and HC2 are derived from ant-EGFR panitumumab. The + symbol indicates the presence of the particular plasmid DNA for transfection, and the − symbol indicates its absence.
FIGURE 5.
FIGURE 5.
Chain drop-out transient transfections to assess the electrostatic steering effect. 2936E cells were transfected with either two or four different plasmid DNAs encoding anti-HER2 trastuzumab and anti-HER2 pertuzumab in which the charged residue pairs in Fab regions were swapped. Six days post-transfection the crude supernatant was loaded in 8–16% Tris-glycine SDS-polyacrylamide gel, subjected to electrophoresis under nonreducing conditions, and Western blotted. The sequence variations V23A, V23B, V23C, and V23D are indicated in Table 4. LC1 and HC1 are derived from anti-HER2 trastuzumab, LC2 and HC2 are derived from ant-HER2 pertuzumab. The + symbol indicates the presence of the particular plasmid DNA for transfection, and the − symbol indicates its absence.
FIGURE 6.
FIGURE 6.
Thermal stability analysis of parental antibodies and anti-HER2 × EGFR hetero-IgG1 variants by differential scanning calorimetry. All antibodies were produced in 2936E cells by transient transfection, purified by protein A, and polished by Superdex 200 size exclusion column. Anti-HER2 trastuzumab IgG1, afucosylated anti-HER2 humAb4D5–8 IgG1, and anti-EGFR E7.6.3 IgG1 were included as internal controls. The sequence variations of anti-HER2 × EGFR hetero-IgG1 variants V12, V23, V24, and V25 are indicated in Table 3. All four anti-HER2 × EGFR hetero-IgG1 have embedded with ADCC enhancement Fc variant W165 (40).
FIGURE 7.
FIGURE 7.
Stable expression of anti-HER2 × EGFR and anti-HER2 × HER2 hetero-IgG1 variants in CHO-K1 cells. A, Western blotting of purified proteins from transient transfection and crude supernatants from stably transfected CHO-K1 cells in SDS-PAGE under nonreducing conditions. B, Western blotting of purified proteins from transient transfection and crude supernatants from stably transfected CHO-K1 cells in SDS-PAGE under reducing conditions. 0.5 μg/lane purified protein from transient transfection and 10 μl/lane crude supernatant were loaded. The crude supernatant as a control from nontransfected CHO-K1 cells was loaded next to the molecular mass standard. Lanes 1–3, anti-HER2 × EGFR hetero-IgG1 V23; lanes 4–6, anti-HER2 × EGFR hetero-IgG1 V23_W165; lanes 7–9, anti-HER2 × HER2 hetero-IgG1 V23; lanes 10–12, anti-HER2 × HER2 hetero-IgG1 V23_W165. Lanes 1, 4, 7, and 10 contained the respective purified protein from transient transfection. Supernatant from two separate CHO-K1 pools was loaded next to the purified protein. C, mass spectrum of intact anti-HER2 × EGFR hetero-IgG1 V23 after deglycosylation by PNGase F. D, mass spectrum of intact anti-HER2 × EGFR hetero-IgG1 V23_W165 after deglycosylation by PNGase F. E, mass spectrum of intact anti-HER2 × HER2 hetero-IgG1 V23 after deglycosylation by PNGase F. F, mass spectrum of intact anti-HER2 × HER2 hetero-IgG1V23_W165 after deglycosylation by PNGase F. G, zoomed in region around the mass at 145,244.32 daltons of anti-HER2 × EGFR hetero-IgG1 V23. Three peaks indicate additional 1 Arg, or 2 Arg, or 3 Arg are retained. H, zoomed in region around the mass at 72,569.73 daltons of anti-HER2 × EGFR hetero-IgG1 V23. ½ Ab of anti-EGFR E7.6.3 (HC2 + LC2) containing additional 1 Arg or 2 Arg and ½ Ab of anti-HER2 humAb4D5–8 (HC1 + LC1) are detectable. I, anti-HER2 humAb4D5–8 HC1 after deglycosylation and reduction. J, anti-EGFR E7.6.3 HC2 after deglycosylation and reduction. K, anti-HER2 humAb4D5–8 LC1 after deglycosylation and reduction. L, anti-EGFR E7.6.3 LC2 after deglycosylation and reduction. Additional 1 Arg, or 2 Arg, or 3 Arg is retained at the C terminus of anti-EGFR E7.6.3 LC2.
FIGURE 8.
FIGURE 8.
Anti-HER2 × EGFR hetero-IgG1 antibodies from stably transfected CHO-K1 cells have comparable binding affinity as parental antibodies. Representative SPR sensorgrams of triplicate injections of 75 nm monomeric rhuEGFR injected at time 0 s followed by 75 nm monomeric rhuHER2 injected at 800 s over the following: A, anti-HER2 × EGFR hetero-IgG1 V23; B, anti-HER2 × EGFR hetero-IgG1 V23_W165; C, anti-EGFR E7.6.3 IgG1; and D, anti-HER2 humAb4D5–8 IgG1. E–G show the SPR sensorgrams (black lines), and the results from nonlinear least squares regression analysis of the data (red lines). Global fits utilize a 1:1 binding model for the triplicate injections of five concentrations of monomeric rhuHER2 ranging between 25.0 and 0.309 nm against captured (E); anti-HER2 × EGFR hetero-IgG1 V23(E); anti-HER2 × EGFR hetero-IgG1 V23_W165 (F); anti-HER2 humAb4D5–8 IgG1 (G). H–J show the SPR sensorgrams (black lines), and the results from nonlinear least squares regression analysis of the data are shown (red lines). Global fits utilize a 1:1 binding model for the triplicate injections of five concentrations of monomeric rhuEGFR ranging between 25.0 and 0.309 nm against captured anti-HER2 × EGFR hetero-IgG1 V23 (H); anti-HER2 × EGFR hetero-IgG1 V23_W165 (I); and anti-EGFR E7.6.3 IgG1 (J).
FIGURE 9.
FIGURE 9.
Anti-HER2 × HER2 hetero-IgG1 antibodies from stably transfected CHO-K1 cells have intermediate binding affinity compared with parental antibodies. SPR sensorgrams (black lines) and the results from nonlinear least squares regression analysis of the data (red lines) are shown. Graphs are the sensorgrams, and the global fits utilize a 1:1 binding model for the triplicate injections of five concentrations of monomeric rhuHER2 ranging between 25.0 and 0.309 nm against captured anti-HER2 × HER2 hetero-IgG1 V23 (A), anti-HER2 × HER2 hetero-IgG1 V23_W165 (B), anti-HER2 humAb4D5–8 IgG1 (C), and anti-HER2 humAb2C4 IgG1 (D).
FIGURE 10.
FIGURE 10.
Hetero-IgG1 antibodies elicit potent ADCC killing to tumor cells and inhibit phosphorylation of molecules in the signaling pathway. A, NCI-N87 as target cells. B, JIMT-1 as target cells. C, SK-BR-3 as target cells. D, BT-474 as target cells. Percent specific lysis was calculated using (RLU values of treated samples subtracted by average RLU value of effector alone) divided by ((the average RLU of untreated cells (effector + target) subtracted by average RLU of effector alone))·100. E, BxPC-3 cells were used for the pEGFR inhibition. F, MCF-7 cells were used for the pHER2 inhibition. G, MCF-7 cells were used for the pHER3 inhibition. H, MCF-7 cells were used for the pAKT inhibition. The level of phosphorylated molecules was detected and analyzed as described under “Experimental Procedures.”
FIGURE 11.
FIGURE 11.
Hetero-IgG1 antibodies increase cellular target internalization levels as compared with the levels mediated by either parental antibody alone or a combination of parental antibodies. Monolayer BxPC-3 cells were exposed to either control human IgG1, anti-HER2 humAb4D5–8 IgG1, anti-EGFR E7.6.3 IgG1, anti-HER2 × EGFR hetero-IgG1 V23 at a final concentration of 5 μg/ml (34 nm) or a combination of anti-HER2 humAb4D5–8 IgG1 and anti-EGFR E7.6.3 IgG1 at a final concentration of 2.5 μg/ml (17 nm) for each antibody. A, time point 0 h. B, anti-HER2 humAb4D5–8 IgG1 at time point 4 h. C, anti-EGFR 7.6.3 IgG1 at time point 4 h. D, combination of anti-HER2 humAb4D5–8 IgG1 and anti-EGFR E7.6.3 IgG1 at time point 4 h. E, anti-HER2 × EGFR hetero-IgG1 V23 at time point 4 h. F, total detectable cell surface binding of antibodies at time point 0 h. G, spot intensity per cell over 4 h of incubation.
FIGURE 12.
FIGURE 12.
Hetero-IgG1 antibodies strongly inhibit tumorigenic growth of human BxPC-3, Panc-1, and Calu-3 cancer cells in mice. A, tumor growth of human BxPC-3 pancreatic cancer cells in female CB-17 SCID mice. The mice were treated once a week for 5 weeks via intraperitoneal administration with either saline, anti-HER2 humAb4D5–8 IgG1 (250 μg), anti-EGFR E7.6.3 IgG1 (250 μg), the combination of anti-HER2 humAb4D5–8 IgG1 (250 μg) with anti-EGFR E7.6.3 IgG1 (250 μg), anti-HER2 × EGFR hetero-IgG1 V23 (500 μg), or anti-HER2 × EGFR hetero-IgG1 V23_W165 (500 μg). B, tumor growth of human Panc-1 pancreatic cancer cells in Rag2−/−/FcγR4−/−/hCD16a+ transgenic mice. The mice were treated once a week for 4 weeks via intraperitoneal administration with either huIgG1 isotype (500 μg), anti-HER2 humAb4D5–8 IgG1with huIgG1 isotype (250 μg each), anti-EGFR E7.6.3. IgG1 with huIgG1 isotype (250 μg each), the combination of anti-HER2 humAb4D5–8 IgG1 with anti-EGFR E7.6.3 IgG1 (250 μg each), anti-HER2 × EGFR heter-IgG1 V23 (500 μg), or anti-HER2 × EGFR heter-IgG1 V23_W165 (500 μg). C, tumor growth of human Calu-3 cancer cells in female NSG mice. The mice were treated once a week for 4 weeks via intraperitoneal administration with either huIgG1 isotype (500 μg), anti-HER2 humAb4D5–8 IgG1 with huIgG1 isotype (250 μg each), anti-HER2 humAb2C4 IgG1 with huIgG1 isotype (250 μg each), the combination of anti-HER2 humAb4D5–8 IgG1 and anti-HER2 humAb2C4 IgG1 (250 μg each), anti-HER2 × HER2 heter-IgG1 V23 (500 μg), or anti-HER2 × HER2 heter-IgG1 V23_W165 (500 μg). The arrows mark the day when antibody was administered. D, summary of statistical analyses for BxPC-3 and Panc-1 xenograft studies.
FIGURE 13.
FIGURE 13.
In vitro stability of anti-HER2 × EGFR and anti-HER2 × HER2 hetero-IgG1 bsAbs in human serum. A, retention of the parent anti-EGFR E7.6.3 IgG1 and anti-HER2 × EGFR hetero-IgG1 V23 when binding to the captured biotin-EGFR. B, retention of the parent anti-HER2 humAb4D5–8 IgG1 and anti-HER2 × EGFR hetero-IgG1 V23 when binding to the captured biotin-HER2. C, retention of the parent anti-HER2 humAb4D5–8 IgG1, parent anti-HER2 humAb2C4 IgG1, and anti-HER2 × HER2 hetero-IgG1 V23 when binding to the captured biotin-HER2. The data were derived from two separate ELISA tests.

Similar articles

See all similar articles

Cited by 31 articles

See all "Cited by" articles
Feedback