Search

During lymphocyte development, the assembly of genes coding for antigen receptors occurs by the combinatorial linking of gene segments. The stochastic nature of this process gives rise to lymphocytes that can recognize self-antigens, thereby having the potential to induce autoimmune disease. Such autoreactive lymphocytes can be silenced by developmental arrest or unresponsiveness (anergy), or can be deleted from the repertoire by cell death. In the thymus, developing T lymphocytes (thymocytes) bearing a T-cell receptor (TCR)-CD3 complex that engages self-antigens are induced to undergo programmed cell death (apoptosis), but the mechanisms ensuring this 'negative selection' are unclear. We now report that thymocytes lacking the pro-apoptotic Bcl-2 family member Bim (also known as Bcl2l11) are refractory to apoptosis induced by TCR-CD3 stimulation. Moreover, in transgenic mice expressing autoreactive TCRs that provoke widespread deletion, Bim deficiency severely impaired thymocyte killing. TCR ligation upregulated Bim expression and promoted interaction of Bim with Bcl-XL, inhibiting its survival function. These findings identify Bim as an essential initiator of apoptosis in thymocyte-negative selection.

Apoptosis is an evolutionarily conserved cell suicide process executed by cysteine proteases (caspases) and regulated by the opposing factions of the Bcl-2 protein family. Mammalian caspase-9 and its activator Apaf-1 were thought to be essential, because mice lacking either of them display neuronal hyperplasia and their lymphocytes and fibroblasts seem resistant to certain apoptotic stimuli. Because Apaf-1 requires cytochrome c to activate caspase-9, and Bcl-2 prevents mitochondrial cytochrome c release, Bcl-2 is widely believed to inhibit apoptosis by safeguarding mitochondrial membrane integrity. Our results suggest a different, broader role, because Bcl-2 overexpression increased lymphocyte numbers in mice and inhibited many apoptotic stimuli, but the absence of Apaf-1 or caspase-9 did not. Caspase activity was still discernible in cells lacking Apaf-1 or caspase-9, and a potent caspase antagonist both inhibited apoptosis and retarded cytochrome c release. We conclude that Bcl-2 regulates a caspase activation programme independently of the cytochrome c/Apaf-1/caspase-9 'apoptosome', which seems to amplify rather than initiate the caspase cascade.

Nature 415, 922–926 (2002). The number of CD4+ CD8+ thymocytes shown in Fig. 1a of this Letter should be × 107, and not × 106 as published. This error does not affect our conclusions.

Messenger RNAs from mouse myeloma cells contain N-6-methyl adenosine and novel 5' termini having 7-methy-guanosine in a 5', 5' triphosphate linkage with ribose-methylated nucleosides. Ten common 5'-terminal sequences of the forms m-7-G-5'-PPP-5'-NmpNp and m-7-G-5'-ppp-5'-NmpNmpNp are reported. Structures like this may be a general feature of mRNA in eukaryotes.

The 5′-terminal sequence of 74 nucleotides reported here does not seem to be involved in the synthesis of phage specific proteins but is probably concerned in the replication of R17 RNA. It is also of comparative interest—the sequence is strikingly similar to the corresponding section of Qβ phage RNA.

Transgenic mice bearing the cellular myc oncogene coupled to the immunoglobulin mu or kappa enhancer frequently develop a fatal lymphoma within a few months of birth. Since the tumours represent represent both immature and mature B lymphocytes, constitutive c-myc expression appears to be highly leukaemogenic at several stages of B-cell maturation. These myc mice should aid study of lymphoma development, B-cell ontogeny and immunoglobulin regulation.

The frequent trisomy of murine chromosome 15 in T lymphomas suggests that it bears one or more genes conducive to T-cell neoplasia1. One such gene seems to be c-myc, the oncogene frequently activated in B-lymphoid tumours either by retroviral insertion, as in the avian bursal lymphomas2,3, or by a translocation to the immunoglobulin heavy-chain locus, as in the predominant t(12; 15) of murine plasmacytomas and the analogous t(14; 8) of human Burkitt lymphomas (for reviews, see refs 1, 4–6). The c-myc gene was strongly implicated in T-cell neoplasia when 15–25% of T lymphomas arising in AKR mice, a strain prone to leukaemia, were found to have retroviral inserts near c-myc7,8. Proviruses near c-myc were also found in several T lymphomas induced by murine leukaemia viruses (MuLV) in both mice9 and rats10, but many of the rat thymomas bear an insert instead at one of several other common sites11–13, at least two of which have murine homologues on chromosome 15 (refs 14, 15). We show here that some murine T lymphomas contain proviral inserts in the recently identified chromosome 15 locus for plasmacytoma variant (6; 15) translocations, which we have denoted pvt-1 (refs 16, 17). Although 6; 15 breakpoints map cytogenetically to the same chromosome band as c-myc18,19, the alterations of pvt-1 in tumours occur at least 72 kilobases (kb) from the c-myc promoters17. The insertions in T lymphomas suggest that an altered pvt-1 locus is conducive to neoplasia in T cells as well as B cells, possibly via long-range effects on c-myc expression.