RAGs found “not guilty”: cleared by DNA evidence

B- and T-cell lymphomas and leukemias are frequent hosts to chromosomal translocations, duplications, and deletions, the best studied of which are known to result in the hyperactivation of oncogenes or the fusion of 2 genes into a chimeric oncogene. How do these DNA-rearranging events occur? For the most part, there are 2 basic requirements: 2 distant double-strand DNA breakages, and joining of the wrong ends together. A fair amount of circumstantial evidence implicates the recombination activating gene (RAG) proteins, RAG-1 and RAG-2, as key suspects in the DNA breakage step for some of these translocations. This is because at the pre-T and pre-B stages of development, RAGs naturally cut the DNA at the T-cell receptor (TCR) and B-cell receptor (BCR or immunoglobulin) loci as part of V(D)J recombination. In this physiologic process, RAGs cleave the DNA adjacent to segments encoding the various parts of the receptor, and the segments are stitched together into a functional receptor gene. The inherent imprecision in this process, in addition to the combinatorial possibilities of the segments that can be used, is the basis for much of the diversity seen in the adaptive immune system. However, pathologic consequences occur when RAGs cut other loci and join them aberrantly. Biochemical studies using purified recombinant RAG protein complexes have shown that their minimal target is the DNA sequence CACA, which occurs millions of times in the typical 6-billion base pair human nucleus. Thus, it has long been suspected that RAGs may be responsible for many of the chromosomal abnormalities in lymphocytes.

H2ax^–/– p53^–/– mice, which typically die from pre-T-cell lymphomas with clonal chromosomal defects, offer an opportunity to test that hypothesis. A significant proportion of the T-cells of H2ax^–/– mice contain transloca-tions, but for the most part they do not de-velop lymphomas. On the other hand, p53^–/– and p53^–/– RAG2^–/– mice die from pre–T-cell lymphomas, but mostly without clonal translocations, deletions, or duplications. Could the chromosomal abnormalities seen in the lymphomas of H2ax^–/– p53^–/– mice be due to RAGs? In this issue of Blood, Bassing and colleagues find that H2ax^–/– p53^–/– RAG2^–/– mice still die from pre-T lymphomas and that these lymphomas still have clonal chromosomal defects. Thus, the RAGs are probably not responsible for these translocations. It would appear that loss of p53 is responsible for the lymphoma and loss of H2ax is responsible for the translocations. So, going back to the translocation paradigm, what is causing the double-strand breaks if not RAGs? And how might H2ax prevent these translocations? These are questions for another study. It is worth noting, however, that in actual patients, a significant proportion of translocations may still be mediated by the RAGs, while in this system loss of H2ax might drastically increase the proportion of translocations by other mechanisms. Still, this study does open the door to the possibility that these other mechanisms may operate to the extent that H2ax—as well as the other components along the same or similar pathways—might fail.