Of humans, frogs, sharks and chickens, which is the odd one out? If you are interested in B cells, the answer is chickens — the first three animals use somatic hypermutation to diversify the variable (V) regions of their rearranged antibody genes, whereas chickens, together with pigs and rabbits, use gene conversion. Two groups have now shown that a putative RNA editing enzyme — activation-induced cytidine deaminase (AID) — essential for somatic hypermutation and immunoglobulin class-switch recombination is also required for gene conversion.

It is unclear why some species use somatic hypermutation and others use gene conversion. The processes are very different — somatic hypermutation involves the introduction of untemplated single base-pair alterations, whereas in gene conversion, new sequences are copied from nearby pseudogenes. However, it has been shown recently that when gene conversion is blocked in a chicken B-cell line, somatic hypermutation can occur instead, indicating that the two processes might be linked mechanistically.

The two new studies made use of the chicken B-cell line DT40, which undergoes gene conversion spontaneously. Hiroshi Arakawa and colleagues, reporting in Science, used a variant of this cell line with a frameshift mutation in its rearranged V segment that does not, therefore, express surface immunoglobulin (IgM). Gene conversion can repair the mutation, leading to the expression of surface IgM. The rate of reversion provides a quick and easy measure of gene conversion. AID was knocked out by gene targeting to create AID−/− DT40 cells. Subclones were expanded for 18 days then analysed for surface IgM expression. The reversion rate in AID−/− cells was reduced by at least 100-fold, and sequencing confirmed that this was due to an absence of gene conversion.

Reuben Harris and colleagues also generated AID−/− cells, but in an IgM+ DT40 cell line. The cells were expanded for 40 days, then IgMlow cells were sorted and their V genes sequenced. None of the 89 AID−/− cells analysed had undergone gene conversion, whereas gene-conversion events were detected in more than one-third of the AID+/+ cells.

These studies promote AID to the position of master controller of antibody gene modifications. The mechanisms of somatic hypermutation, gene conversion and class-switch recombination are unknown; however, it has been proposed that DNA breaks (double or single stranded) might be pivotal to all three processes. Harris et al. conclude by predicting, “it is likely that AID is involved in the formation of an initiating DNA lesion common to switch recombination, hypermutation and gene conversion”.