Humanization of Monoclonal Antibodies

HUMANIZATION OF MONOCLONAL ANTIBODIES

Monoclonal antibodies could theoretically be used as magic bullets to kill human cancer cells by aiming them at specific molecules appearing only on the surface of cancer cells. Ironically, the main problem is that the human immune system regards antibodies from mice or other animals as foreign molecules themselves, and so attempts to destroy them!

One approach that may partly solve this problem is using genetic engineering to make humanized monoclonals (Fig. 6.9A). This takes advantage of the fact that only the variable or V-region of the antibody recognizes the antigen. The constant or C-region may therefore be replaced. The first-generation hybridoma is isolated and cultured, generally using mouse B cells. Then the DNA encoding the mouse monoclonal antibody is isolated and cloned. The DNA for the constant region of the mouse antibody is replaced with the corresponding human DNA sequence. The V region is left alone. The human/mouse hybrid gene is then put back into a second mouse myeloma cell for production of antibody in culture. Although not fully human, the hybrid is less mouselike and provokes much less reaction from the human immune system.

Further humanization can be accomplished by altering those parts of the V region that are not directly involved in binding the antigen. A closer look at the V-region of each chain shows that most of the variation is restricted to three short segments that form loops on the surface of the antibody, thus forming the antigen binding site (see Fig. 6.9B). These are known as hypervariable regions or as complementarity determining regions (CDRs). Overall, each antigen binding site consists of six CDRs, three from the light chain and three from the heavy chain. Full humanization of an antibody involves cutting out the coding regions for these six CDRs from the original antibody and splicing them into the genes for human light and heavy chains.