Seeking high-priority mutations enabling successful antibody-breeding: systematic analysis of a mutant that gained over 100-fold enhanced affinity

“Antibody-breeding” has provided therapeutic/diagnostic antibody mutants with greater performance than native antibodies. Typically, random point mutations are introduced into the V(H) and V(L) domains of parent antibodies to generate diverse libraries of single-chain Fv fragments (scFvs), from which evolved mutants are selected. We produced an scFv against estradiol-17β with 11 amino acid substitutions and a >100-fold improved affinity constant (K(a) = 1.19 × 10(10) M(−1)) over the parent scFv, enabling immunoassays with >30-fold higher sensitivity. We systematically analyzed contributions of these substitutions to the affinity enhancement. Comparing various partial scFv revertants based on their K(a)s indicated that a revertant with four substitutions (V(H)-L100gQ, V(L)-I29V, -L36M, -S77G) exhibited somewhat higher affinity (K(a) = 1.46 × 10(10) M(−1)). Finally, the V(H)-L100gQ substitution, occurring in V(H) complementarity-determining region (CDR) 3, was found to be the highest-priority for improving the affinity, and V(L)-I29V and/or V(L)-L36M cooperated significantly. These findings encouraged us to reconsider the potential of V(H)-CDR3-targeting mutagenesis, which has been frequently attempted. The substitution(s) wherein might enable a “high rate of return” in terms of selecting mutants with dramatically enhanced affinities. The “high risk” of generating a tremendous excess of “junk mutants” can be overcome with the efficient selection systems that we developed.