A new approach for generating bispecific antibodies based on a common light chain format and the stable architecture of human immunoglobulin G1

Abstract

Bispecific antibodies combine two different antigen-binding sites in a single molecule, enabling more specific targeting, novel mechanisms of action, and higher clinical efficacies. Although they have the potential to outperform conventional monoclonal antibodies, many bispecific antibodies have issues regarding production, stability, and pharmacokinetic properties. Here, we describe a new approach for generating bispecific antibodies using a common light chain format and exploiting the stable architecture of human immunoglobulin G₁ We used iterative experimental validation and computational modeling to identify multiple Fc variant pairs that drive efficient heterodimerization of the antibody heavy chains. Accelerated stability studies enabled selection of one Fc variant pair dubbed "DEKK" consisting of substitutions L351D and L368E in one heavy chain combined with L351K and T366K in the other. Solving the crystal structure of the DEKK Fc region at a resolution of 2.3 Å enabled detailed analysis of the interactions inducing CH3 interface heterodimerization. Local shifts in the IgG backbone accommodate the introduction of lysine side chains that form stabilizing salt-bridge interactions with substituted and native residues in the opposite chain. Overall, the CH3 domain adapted to these shifts at the interface, yielding a stable Fc conformation very similar to that in wild-type IgG. Using the DEKK format, we generated the bispecific antibody MCLA-128, targeting human EGF receptors 2 and 3. MCLA-128 could be readily produced and purified at industrial scale with a standard mammalian cell culture platform and a routine purification protocol. Long-term accelerated stability assays confirmed that MCLA-128 is highly stable and has excellent biophysical characteristics.

Keywords: antibody; antibody engineering; anticancer drug; crystal structure; immunoglobulin G (IgG).

Figure 1.

Identification of CH3 residues that block IgG homodimer formation. Constructs contain CH3 substitutions as indicated. Constructs containing CH3 interfaces that allow homodimer formation are visible as bands at 144 kDa, whereas interference with dimer formation is demonstrated by the presence of 74-kDa bands. Wt denotes wild-type IgG₁.

Figure 2.

Efficiency of bispecific formation by DEKK. Transient transfections were performed using skewed DNA ratios of two vectors each expressing one arm of the DEKK Fc variant pair, followed by nMS analysis to determine the relative amounts of the IgG species (A–C). As a control, transfections were performed using two different IgG₁ heavy chains containing WT Fc regions (D). Black and gray bars represent duplicate experiments using different VH gene combinations in the context of the DEKK Fc variant pair or WT Fc regions. The VH gene combinations used were MF1337 × MF1122 and MF1337 × MF2729. A, DE/KK DNA transfection ratio 5:1; B, DE/KK DNA transfection ratio 1:1; C, DE/KK DNA transfection ratio 1:5. DE indicates L351D,L368E; KK indicates T366K,L351K. DEDE, KKKK, DEKK, DE, and KK denote the homo- and heterodimers and “half-IgGs” of the L351D,L368E and T366K,L351K arms. D, WT Fc regions, DNA transfection ratio 1:1. A and B denote heavy chains containing different VH regions. AB indicates bispecific molecules, and AA and BB indicate monospecific antibodies.

Figure 3.

Crystal structure of the Fc region of DEKK at 2.3-Å resolution. A, structural superimposition of the DEKK and WT Fc regions (PDB code 1L6X (14)) based on CH3 domains. DEKK chain A (T366K,L351K) is colored orange; chain B (L351D,L368E) is blue, and the Fc-III peptide is yellow; the WT structure is gray. B, electron density (2F_obs − F_calc) at 1σ contour level of DEKK residues in chain A (orange) and in chain B (blue); these residues are annotated in C. C, view of the interactions within the CH3 interface of the engineered residues in DEKK (shown as sticks). The dashed lines indicate the distances between interacting atoms in Å. Color scheme as in A. D, bottom view of the CH3/CH3 dimerization interface of the superimposed structures of the DEKK and WT Fc regions (PDB code 1L6X). The dashed lines measure the distances between residues 351(A)-(B) and 353(A)-(B) in both DEKK and WT as provided under “Results.” The arrows indicate the conformational rearrangement of the helix consisting of residues 354–359 in DEKK compared with the same helix in the WT molecule. Color scheme as in A. E, view of the CH3 interface of DEKK superimposed to the four best models generated by HADDOCK (light blue). Color scheme for DEKK as in A.

Figure 4.

Crystal structure of the Fc region of DEDE at 2.2-Å resolution. A, structural superimposition of DEDE and WT Fc regions (PDB code 1L6X (14)), based on CH3 domains. DEDE is colored cyan; WT is gray. B, zoomed-in bottom view of the CH3/CH3 dimerization interface of the superimposed structures of DEDE and WT Fc regions (PDB code 1L6X), based on chain A. Mutated residues are shown as sticks. The dashed lines indicate the distance between residues 351(A)-(B) as provided under “Results.”

Figure 5.

CEX was used to remove product-related impurities from a protein A-purified sample of MCLA-128 produced using the DEKK Fc region pair. A, analytical CEX chromatogram of the preparative CEX column load shows early eluting DEDE homodimers running at around 12 min, bispecific antibody MCLA-128 eluting between 14 and 20.5 min, and late eluting KK half-IgG running at around 23 min. B, analytical CEX chromatogram of the preparative CEX column eluate showing only bispecific antibody MCLA-128, demonstrating efficient removal of the DEDE and KK impurities. DEDE denotes the homodimer of the L351D,L368E Fc region variant, and KK denotes half-IgGs of the T366K,L351K Fc region variant.