Specificity improvement of a recombinant anti-testosterone Fab fragment by CDRIII mutagenesis and phage display selection.

The monoclonal antibodies so far developed by hybridoma technology have not had high enough specificity or affinity to distinguish the closely related steroid hormones in routine clinical assays. We have employed random mutagenesis and phage display approaches to improve the specificity of one anti-testosterone monoclonal antibody (3-C4F5). The affinity of the antibody is 0.3 x 10(9) M(-1) and the cross-reactivities with most of the related steroids are low. However, the antibody cross-reacts about 1% with dehydroepiandrosterone sulfate (DHEAS) and owing to the high DHEAS serum concentration this is about 1000-fold too high for clinical immunoassays. The complementarity-determining regions (CDRs) of the heavy and light chains, which were predicted by molecular modelling to be in close contact with the testosterone (TES) ligand, were randomized and mutant Fab libraries were cloned into a phagemid vector. Binders were selected by a competitive panning procedure. By combining the identified light and heavy chain CDRIII mutations the TES affinity was preserved at the wild-type level but DHEAS cross-reactivity was decreased to 0.03%. An important finding was that by the competitive panning procedure the overall binding specificity of the 3-C4F5 antibody was refined, since the cross-reactivities to related steroids were also significantly decreased in the combined mutant.