A Common Ancestry for Multiple Catalytic Antibodies Generated Against a Single Transition-state Analog

Immunization with a single haptenic transition-state analog generates a few catalytic antibodies among the dozens of antibodies capable of binding the hapten. The diversity of the immune response has raised some fundamental issues, such as How do catalytic and noncatalytic antibodies differ on a structural basis? To address this issue, the variable region primary sequences of 11 antibodies (including 6 catalytic and 5 noncatalytic antibodies) elicited against a single haptenic transition-state analog were deduced from cDNA sequences. Cluster analyses using phylogenetic trees constructed by the neighbor-joining method have revealed that the amino acid sequences of noncatalytic antibodies bear no relationship to one another, while the catalytic antibodies share significant structural identity. Furthermore, no catalytic antibodies possessing amino acid sequences with high homology to those of noncatalytic antibodies were detected. Five catalytic antibodies examined showed 89-95% and 74-84% sequence homologies in the complete light- and heavy-chain variable regions, respectively. Thus, it seems likely that the catalytic antibodies elicited against a single hapten use the canonical set of variable region genes. Interestingly, one catalytic antibody showed only limited sequence similarity to the other catalytic antibodies and was found to exhibit a distinctly different substrate specificity. From the broad range of their binding constants to the hapten, it is unlikely that highly homologous catalytic antibodies are generated as a result of simple high-affinity choices. These results emphasize the utility of rationally designed transition-state analogs for the induction of antibody molecules with catalytic activity.