Antibody Evolution and Repertoire Development

Sammanfattning: Antibodies are key players of the immune system in higher vertebrates, which provide a defense against potentially lethal threats from the environment. Besides their importance in the immune defense, antibodies have a great potential as reagents in biological chemistry and diagnostics, and as therapeutic agents against both infectious diseases and cancer. Despite tremendous advances in all fields regarding antibodies, the understanding of the processes that shape antibody responses in vivo is far from complete. In this thesis, which is based on five original papers, I present work that deals with aspects of the evolution of antibodies and the development of antibody repertoires. By analyzing the inherited repertoires of human genes encoding the variable domains of antibodies, I have discovered regions with accumulations of repetitive trinucleotide motifs. These repeats were mainly found in the complementarity determining regions (CDR), and most likely target them with insertions and deletions during the maturation of an antibody response. In order to investigate the functional consequences of such modifications, insertions and deletions were introduced into the CDR of antibody fragments by molecular engineering. The results showed that modifications of this kind are well tolerated in the CDR of the heavy chain, and can be utilized to expand sequence and structure space of an antigen-binding site beyond what is encoded by the germline gene repertoire. Thus, insertions and deletions seem to be an efficient way of expanding antibody sequence and structure space both in vivo and in vitro, and it is conceivable that these modifications can also be used in antibody engineering to create antibodies against specific targets. I have also studied the development of antibodies reactive with a weakly immunogenic epitope on human cytomegalovirus, and what the critical parameters are that determine the neutralizing capacity of such antibodies. The results from these studies showed that the genes encoding one of the two existing human antibodies reactive with this epitope do not have the intrinsic features required for high-affinity interaction with the epitope, as mutations are required at key positions. The poor immunogenicity of this epitope may in fact be a consequence of the lack of an imprinted specificity in the human germline repertoire. Furthermore, the obtained data demonstrated that a divalent antibody format is required for effective virus neutralization via this epitope, and that the neutralizing potential of a repertoire of antibodies reactive with this epitope is determined by the reaction rate kinetics and fine-specificity of the interaction with the epitope. These results provide insights into the development of human antibody repertoires against weakly immunogenic targets, and are also discussed in the context of vaccine development.