Advances in Antibody Design

Abstract

The use of monoclonal antibodies as therapeutics requires optimizing several of their key attributes. These include binding affinity and specificity, folding stability, solubility, pharmacokinetics, effector functions, and compatibility with the attachment of additional antibody domains (bispecific antibodies) and cytotoxic drugs (antibody-drug conjugates). Addressing these and other challenges requires the use of systematic design methods that complement powerful immunization and in vitro screening methods. We review advances in designing the binding loops, scaffolds, domain interfaces, constant regions, post-translational and chemical modifications, and bispecific architectures of antibodies and fragments thereof to improve their bioactivity. We also highlight unmet challenges in antibody design that must be overcome to generate potent antibody therapeutics.

Keywords: CDR; Fab; IgG; VH; complementarity-determining region; scFv.

Figure 1

Molecular architecture of an immunoglobulin G1 (IgG1) antibody. An IgG consists of two heavy chains (blue) and two light chains ( pink). (Left) The crystal structure of an antigen-binding fragment (Fab; Protein Data Bank identification number, 3NZ8). The Fab is composed of variable heavy (V_H) and light (V_L) domains as well as two constant domains (C_H1 and C_L). Each variable domain displays three binding loops (complementarity-determining regions, CDRs), which mediate antigen recognition. The CDRs in the V_H domain are denoted as H1, H2, and H3 (blue); the CDRs in the V_L domain are denoted as L1, L2, and L3 ( pink). (Right) The crystallizable fragment (Fc; Protein Data Bank identification number, 1E4K) contains two constant domains (C_H2 and C_H3) as well as glycans in the C_H2 domain ( green). The Fc fragment mediates antibody effector function.

Figure 2

Key attributes of antibodies that must be collectively optimized to generate effective immunoglobulins for different applications. A key challenge is that optimizing one property can lead to deleterious impacts on others.

Figure 3

Nature-inspired design and evolution of phospho-specific antibodies. This approach uses natural anion-binding motifs within the complementarity-determining regions (CDRs) of antibodies to generate libraries for identifying antibodies specific for phosphoserine, phosphotyrosine, and phosphothreonine (20). The first round of randomization and selection yields antibodies with anion-binding motifs specific for each type of modification, and the second round identifies antibodies with heavy chain CDR3 (HCDR3) and light chain CDR3 (LCDR3) loops that are specific for different phosphopeptides. Figure redrawn from Reference .

Figure 4

Mutations identified using structure-based and related design methods that enhance the conformational (folding) stability of antibody fragments. Four mutations were introduced into the variable domains of a poorly stable antibody fragment (red ) that generated a more stable one ( green) (44). The crystal structures were obtained from the Protein Data Bank: (top) 3HC0 and (bottom) 3HC4. Abbreviations: λ_em, average (center of mass) fluorescence emission wavelength. Data from Reference .

Figure 5

Design methods for increasing antibody solubility. Multiple approaches were found to increase the solubility of a glycoprotein LINGO-1 antibody and to have little impact on binding affinity. Data from Reference . Abbreviations: CDR, complementarity-determining region; IgG, immunoglobulin G.

Figure 6

Structure-guided design and selection of crystallizable fragment (Fc) mutations that increase complement-dependent cytotoxicity (CDC). (a) Residues in the heavy chain constant domain C_H2 (Protein Data Bank identification number, 1E4K) that form the putative C1q binding center. (b) Mutations identified using structure-based methods that increase CDC (85). Evaluation of (c) CDC and (d) antibody-dependent cell-mediated cytotoxicity (ADCC) of wild-type and three mutants of an immunoglobulin G1 (IgG1) anti-CD20 antibody. Data from Reference .

Figure 7

Design and evaluation of the bioactivity of antibodies conjugated at different sites with a cytotoxic drug. (a) The sites mutated to cysteine are highlighted in the crystal structures of the antigen-binding fragment (Fab) and crystallizable fragment (Fc), and (b) these sites show a range of solvent accessibilities. (c) Evaluation of the clearance rates of antibodies injected into mice. Although the total antibody levels were similar for injections of each antibody–drug conjugate (ADC), the fraction of intact ADC was highest for the light chain (LC) variant V205C and lowest for the Fc S396C variant. (d ) Mice dosed equally with each ADC showed significant differences in survival, and these differences were consistent with the fraction of intact ADC. Abbreviation: HC, heavy chain. Figure adapted with permission from Macmillan Publishers Ltd.: Nature Biotechnology (109), copyright 2012.

Figure 8

Molecular architectures of bispecific monoclonal antibodies (mAbs). Two mAbs are recombined into different bispecific architectures. A quadroma Triomab (Trion Pharma, Munich, Germany) comprises one heavy chain–light chain pair of a rat immunoglobulin G2 (IgG2) and one heavy chain–light chain pair of a murine IgG2 antibody (128). The knobs-into-holes architecture consists of an opposing cavity and protrusion in the heavy chain constant C_H3 domains to enforce the heteropairing of heavy chains (127). The CrossMAb (Roche, Basel, Switzerland) architecture involves swapping the light chain constant C_L and the heavy chain constant C_H1 domains onto opposite chains to enforce correct light-chain pairing, and also uses knobs-into-holes mutations to enforce correct heavy-chain pairing (140). Dual-variable-domain antibodies have the variable domains from one antibody added to the N terminus of the heavy and light chains of the other antibody (142). IgG–scFv (single-chain variable fragment) bispecific antibodies contain the variable domains of one antibody—which are reformatted as an scFv—fused to the terminus of the heavy or light chains of the second antibody (132).