AID: a riddle wrapped in a mystery inside an enigma.

To combat the ever-changing pool of pathogens we face, B cells generate highly optimized antibodies in two distinct steps. A large variety of antibodies are first generated randomly by V(D)J recombination, and then, upon encountering an antigen, antibodies are fine-tuned by somatic hypermutation and class switch recombination--both of which are initiated by the same protein, activation-induced cytidine deaminase (AID). All three processes are highly mutagenic, and mistargeting of each of these has been shown to contribute to tumorigenesis. We study these processes because they provide an excellent model to understand how highly mutagenic reactions are channeled into productive use by cells and the consequent risk this carries. In this review, we will discuss many of the outstanding questions in the field that we grapple with while developing a consistent model for AID action. We will also discuss the complexity added to these models by the recent finding that AID might be part of a demethylase complex.