Structural correspondences between mouse and human immunoglobulin VH genes. Application to the humanization of mouse monoclonal antibodies

A high-affinity macaque antibody Fab with human-like framework regions obtained from a small phage display immune library

Tetanus toxoid was used as a model antigen for the immunization of Macaca fascicularis. Using peripheral blood as a template, specific Fab-encoding genes were amplified by PCR on the fourth day after the final boost, and cloned in a phagemidic vector (pComb3X) as a small immune library (5 x 10(5) clones). A high-affinity Fab (Kd = 4 x 10(-10) M), 6-ATT, was isolated from this library by panning. The genes encoding 6-ATT were found to be similar to human immunoglobulin germline genes, and were assigned to subgroups of human V, (D) or J genes by IMGT/V-QUEST. Overall, the Fab variable domain framework regions were 93% identical to the representative genes and alleles of the human subgroups, this level of identity being similar to that between genes of the same human subgroup. This strategy could be used for the isolation of high-affinity, human-like Fab fragments directed against numerous antigens.

The differences between the structural repertoires of VH germ-line gene segments of mice and humans: implication for the molecular mechanism of the immune response

Although human and murine antibodies are similar when considering their diversification strategies, they differ in the proportion by which kappa and lambda type chains are present in their receptive V, repertoires. It has been shown that this difference implies a divergence in the structural repertoire of the kappa and lambda genes of these species. Nonetheless, the differences in VH have not been systematically studied. In this paper a systematic characterization of the VH structural repertoire of mice is made, so that a comparison with the VH structural repertoire of humans, described in detail elsewhere, could be properly accomplished. Our study shows the structural repertoire of mice to be dominated by canonical structure class 1-2 (approximately 60%), while in humans the dominant one is class 1-3 (approximately 40%). Analysis of the evolutionary relationships between human and mice suggest that this divergence may have a functional meaning. The implications of such findings are discussed.