No abstract

DNA double-strand breaks (DSBs) introduced in the switch (S) regions are intermediates during immunoglobulin class switch recombination (CSR). These breaks are subsequently recognized, processed, and joined, leading to recombination of the two S regions. Nonhomologous end-joining (NHEJ) is believed to be the principle mechanism involved in DSB repair during CSR. One important component in NHEJ, Artemis, has however been considered to be dispensable for effi cient CSR. In this study, we have characterized the S recombinational junctions from Artemis-defi cient human B cells. S -S ␣ junctions could be amplifi ed from all patients tested and were characterized by a complete lack of " direct " end-joining and a remarkable shift in the use of an alternative, microhomology-based endjoining pathway. S -S ␥ junctions could only be amplifi ed from one patient who carries " hypomorphic " mutations. Although these S -S ␥ junctions appear to be normal, a significant increase of an unusual type of sequential switching from immunoglobulin (Ig)M, through one IgG subclass, to a different IgG subclass was observed, and the S ␥ -S ␥ junctions showed long microhomologies. Thus, when the function of Artemis is impaired, varying modes of CSR junction resolution may be used for different S regions. Our fi ndings strongly link Artemis to the predominant NHEJ pathway during CSR.

Immunoglobulin class switch recombination (CSR) is initiated by a B-cell-specific factor, activationinduced deaminase, probably through deamination of deoxycytidine residues within the switch (S) regions. The initial lesions in the S regions are subsequently processed, resulting in the production of DNA double-strand breaks (DSBs). These breaks will then be recognized, edited and repaired, finally leading to the recombination of the two S regions. Two major repair pathways have been implicated in CSR, the predominant non-homologous end joining (NHEJ) and the alternative endjoining (A-EJ) pathways. The former requires not only components of the 'classical' NHEJ machinery, i.e. Ku70/Ku80, DNA-dependent protein kinase catalytic subunit, DNA ligase IV and XRCC4, but also a number of DNA-damage sensors or adaptors, such as ataxia-telangiectasia mutated, gH2AX, 53BP1, MDC1, the Mre11-Rad50-NBS1 complex and the ataxia telangiectasia and Rad3-related protein (ATR). The latter pathway is not well characterized yet and probably requires microhomologies. In this review, we will focus on the current knowledge of the predominant NHEJ pathway in CSR and will also give a perspective on the A-EJ pathway.