V(D)J recombination is a site-specific event that generates the immune repertoire in mammalian lymphocytes (1). This event depends on the successful execution of a series of steps by protein factors, some of which function exclusively during V(D)J recombination, whereas others are also essential for doublestrand break (DSB) repair (2). Rag1, Rag2, and terminal deoxynucleotidyl transferase (TdT) are lymphoid specific and function only during V(D)J recombination, whereas Ku70, Ku86, DNA-dependent protein kinase catalytic subunit (DNAPKcs), XRCC4, and ligase IV are essential for both V(D)J recombination and DSB repair. Some of these factors interact with each other, suggesting a mechanism by which proteins are recruited to sites of recombination (3, 4).TdT adds nucleotides in a template-independent manner at coding ends during recombination of Ig and T cell receptor gene segments, thereby expanding the diversity of antigen receptors. It exists as a 58-kDa protein, although smaller products of 55, 44, 32, and 12 kDa have been detected in cell extracts (5). The N terminus of TdT contains a conserved BRCT-like (BRCA1 C-terminal) sequence, a protein-protein interaction domain (6, 7) that was first identified in the breast cancer suppressor gene BRCA1 and subsequently found in 50 other proteins, some of which function in DNA repair and recombination (8,9).The Ku autoantigen is a heterodimer composed of 70-and 86-kDa subunits that form the DNA binding component of an associated 460-kDa DNA-PKcs (3). Ku86-deficient mice are able to initiate V(D)J recombination, but the intermediates of the cleavage reaction are not processed, resulting in accumulation of hairpin coding ends and blunt signal ends (10). The essential role of Ku in this process led us to ask whether it interacts directly with TdT, which modifies these intermediates by the insertion of nucleotides. Notably, the residual coding joints present in Ku86 Ϫ͞Ϫ mice were found to be devoid of N regions, suggesting that the Ku86 protein might regulate TdT activity in vivo (11). We have tested whether Ku and TdT interact directly in vitro and whether such interactions also occur in intact cells treated with a DNA-damaging agent. Materials and MethodsCell Culture. The human Molt-4 (pre-T cell) lymphoblast cell line was maintained at 37°C in RPMI medium 1640 supplemented with 10% (vol͞vol) FBS) and penicillin͞streptomycin. Cos-7 cells were maintained at 37°C in DMEM supplemented with 10% (vol͞vol) FBS. The Sf9 insect cell line was maintained at 27°C in Grace's insect medium supplemented with 10% (vol͞vol) FBS and penicillin͞streptomycin.

No abstract

No abstract

DNA replication in the nucleus of eukaryotic cells is restricted to the S phase of the cell cycle, and different genes are duplicated at specific times, according to a well-defined temporal order. We have investigated whether activation of initiation sites, in proximity to genes that are replicated in different portions of the S phase, could be detected when synchronized 10T1/2 cells were maintained in aphidicolin (APC), an inhibitor of DNA polymerases alpha and delta. Cells released from confluence arrest into medium containing 2 micrograms/mL APC progressed into the S phase, and nascent DNA accumulated during incubations of 24 and 32 h. Exposure to APC for 40 or 48 h resulted in growth of the radiolabeled DNA into larger molecules. Replicating DNA was isolated in CsCl gradients and probed with 32P-labeled gene probes for early-replicating genes (e.g., Ha-ras, mos, and myc) and a late-replicating gene (VH Ig). DNA replicated during the 24-h incubation in APC was enriched in Ha-ras gene sequences. The VH Ig gene did not replicate in cells incubated for as long as 56 h with APC. The myc and the mos genes were detected after 32 and 40 h in APC, respectively. The myc gene is replicated in 10T1/2 cells after Ha-ras but before mos. Therefore, the order of activation of these genes was conserved in the presence of APC. The delay in replication of myc and mos correlated well with the slowing of DNA replication by APC.

Control of initiation of transcription of the human terminal deoxynucleotidyl transferase (TdT) gene was investigated by using an in vitro transcription assay. The precise contribution of discrete basal promoter elements to transcription initiation was determined by testing deletion and substitution mutations. The primary element, contained within the region spanning -34 to -14 bp relative to the transcription start site, accounted for 80% of basal promoter activity. TdT promoter activity required the sequence ACCCT at -24 to -20 bp since a dramatic decrease in transcription initiation was observed after mutation of this sequence, whereas mutation of the adjacent sequence from -32 to -25 bp did not alter promoter activity. The secondary element contained sequences surrounding the transcription start site and had 20% of promoter activity. Deletion of both elements completely abolished transcription initiation. Initiator characteristics of the secondary element were revealed by using the in vitro assay: promoter sequences at the transcription start site were sufficient to direct accurate initiation at a single site. Mutation of the sequence GGGTG spanning the transcription start site resulted in loss of transcription initiation. Both the primary and secondary elements were nonhomologous to corresponding regions from the mouse TdT gene promoter. While the human basal promoter functioned in the absence of TATA consensus sequences or GC-rich SP1 binding sites, it was dependent on active TFIID. In contrast to other TATA-less promoters, purified TATA binding protein substituted for the TFIID complex and restored promoter activity to TFIID-inactivated nuclear extracts.