DNA ligase IV is essential for V(D)J recombination and DNA double-strand break repair in human precursor lymphocytes

Specific interaction of IP6 with human Ku70/80, the DNA-binding subunit of DNA-PK

In eukaryotic cells, DNA double-strand breaks can be repaired by non-homologous end-joining, a process dependent upon Ku70/80, XRCC4 and DNA ligase IV. In mammals, this process also requires DNA-PK(cs), the catalytic subunit of the DNA-dependent protein kinase DNA-PK. Previously, inositol hexakisphosphate (IP6) was shown to be bound by DNA-PK and to stimulate DNA-PK-dependent end-joining in vitro. Here, we localize IP6 binding to the Ku70/80 subunits of DNA- PK, and show that DNA-PK(cs) alone exhibits no detectable affinity for IP6. Moreover, proteolysis mapping of Ku70/80 in the presence and absence of IP6 indicates that binding alters the conformation of the Ku70/80 heterodimer. The yeast homologue of Ku70/80, yKu70/80, fails to bind IP6, indicating that the function of IP6 in non-homologous end-joining, like that of DNA-PK(cs), is unique to the mammalian end-joining process.

Role of poly(ADP-ribosyl)ation in DNA-PKcs- independent V(D)J recombination

V(D)J recombination is critical to the generation of a functional immune system. Intrinsic to the assembly of antigen receptor genes is the formation of endogenous DNA double-strand breaks, which normally are excluded from the cellular surveillance machinery because of their sequestration in a synaptic complex and/or rapid resolution. In cells deficient in double-strand break repair, such recombination-induced breaks fail to be joined promptly and therefore are at risk of being recognized as DNA damage. Poly(ADP-ribose) polymerase-1 is an important factor in the maintenance of genomic integrity and is believed to play a central role in DNA repair. Here we provide visual evidence that in a recombination inducible severe combined immunodeficient cell line poly(ADP-ribose) formation occurs during the resolution stage of V(D)J recombination where nascent opened coding ends are generated. Poly(ADP-ribose) formation appears to facilitate coding end resolution. Furthermore, formation of Mre11 foci coincide with these areas of poly(ADP-ribosyl)ation. In contrast, such a response is not observed in wild-type cells possessing a functional catalytic subunit of DNA-dependent protein kinase (DNA-PK(cs)). Thus, V(D)J recombination invokes a DNA damage response in cells lacking DNA-PK(cs) activity, which in turn promotes DNA-PK(cs)-independent resolution of recombination intermediates.

The V(D)J recombinase efficiently cleaves and transposes signal joints

V(D)J recombination generates two types of products: coding joints, which constitute the rearranged variable regions of antigen receptor genes, and signal joints, which often form on immunologically irrelevant, excised circular molecules that are lost during cell division. It has been widely believed that signal joints simply convert reactive broken DNA ends into safe, inert products. Yet two curious in vivo observations made us question this assumption: signal ends are far more abundant than coding ends, and signal joints form only after RAG expression is downregulated. In fact, we find that signal joints are not at all inert; they are cleaved quite efficiently in vivo and in vitro by a nick-nick mechanism and form an excellent substrate for RAG-mediated transposition in vitro, possibly explaining how genomic stability in lymphocytes may be compromised.

DNA ligase I null mouse cells show normal DNA repair activity but altered DNA replication and reduced genome stability

DNA ligase I is the key ligase for DNA replication in mammalian cells and has also been reported to be involved in a number of recombination and repair processes. Our previous finding that Lig1 knockout mouse embryos developed normally to mid-term before succumbing to a specific haematopoietic defect was difficult to reconcile with a report that DNA ligase I is essential for the viability of cultured mammalian cells. To address this issue, we generated a second Lig1 targeted allele and found that the phenotypes of our two Lig1 mutant mouse lines are identical. Widely different levels of Lig1 fusion transcripts were detected from the two targeted alleles, but we could not detect any DNA ligase I protein, and we believe both are effective Lig1 null alleles. Using foetal liver cells to repopulate the haematopoietic system of lethally irradiated adult mice, we demonstrate that the haematopoietic defect in DNA-ligase-I-deficient embryos is a quantitative deficiency relating to reduced proliferation rather than a qualitative block in any haematopoietic lineage. DNA ligase I null fibroblasts from Lig1 mutant embryos showed an accumulation of DNA replication intermediates and increased genome instability. In the absence of a demonstrable deficiency in DNA repair we postulate that, unusually, genome instability may result directly from the DNA replication defect. Lig1null mouse cells performed better in the survival and replication assays than a human LIG1 point mutant, and we suggest that the complete absence of DNA ligase I may make it easier for another ligase to compensate for DNA ligase I deficiency.

The mechanism and regulation of chromosomal V(D)J recombination

V(D)J recombination is of fundamental importance to the generation of diverse antigen receptor repertoires. We review our current understanding of the V(D)J recombination reaction and how it is regulated during lymphocyte development. We also discuss how defects in the mechanism or regulation of V(D)J recombination can lead to human disease.

Binding of inositol hexakisphosphate (IP6) to Ku but not to DNA-PKcs

The nonhomologous DNA end joining (NHEJ) pathway is responsible for repairing a major fraction of double strand DNA breaks in somatic cells of all multicellular eukaryotes. As an indispensable protein in the NHEJ pathway, Ku has been hypothesized to be the first protein to bind at the DNA ends generated at a double strand break being repaired by this pathway. When bound to a DNA end, Ku improves the affinity of another DNA end-binding protein, DNA-PK(cs), to that end. The Ku.DNA-PK(cs) complex is often termed the DNA-PK holoenzyme. It was recently shown that myo-inositol hexakisphosphate (IP(6)) stimulates the joining of complementary DNA ends in a cell free system. Moreover, the binding data suggested that IP(6) bound to DNA-PK(cs) (not to Ku). Here we clearly show that, in fact, IP(6) associates not with DNA-PK(cs), but rather with Ku. Furthermore, the binding of DNA ends and IP(6) to Ku are independent of each other. The possible relationship between inositol phosphate metabolism and DNA repair is discussed in light of these findings.

Defining interactions between DNA-PK and ligase IV/XRCC4

Non-homologous end joining (NHEJ) is a major pathway for the repair of DNA double-strand breaks (DSBs) in mammalian cells. DNA-dependent protein kinase (DNA-PK), ligase IV, and XRCC4 are all critical components of the NHEJ repair pathway. DNA-PK is composed of a heterodimeric DNA-binding component, Ku, and a large catalytic subunit, DNA-PKcs. Ligase IV and XRCC4 associate to form a multimeric complex that is also essential for NHEJ. DNA-PK and ligase IV/XRCC4 interact at DNA termini which results in stimulated ligase activity. Here, we define interactions between the components of these two essential complexes, DNA-PK and ligase IV/XRCC4. We find that ligase IV/XRCC4 associates with DNA-PK in a DNA-independent manner. The specific protein-protein interactions that mediate the interaction between these two complexes are further identified. Direct interactions between ligase IV and Ku as well as between XRCC4 and DNA-PKcs are shown. In contrast, binding of ligase IV to DNA-PKcs or XRCC4 to Ku is very weak or non-existent. Our data defines the specific protein pairs involved in the association of DNA-PK and ligase IV/XRCC4, and suggests a molecular mechanism for coordinating the assembly of the DNA repair complex at DNA breaks.

Hairpin opening and overhang processing by an Artemis/DNA-dependent protein kinase complex in nonhomologous end joining and V(D)J recombination

Mutations in the Artemis protein in humans result in hypersensitivity to DNA double-strand break-inducing agents and absence of B and T lymphocytes (radiosensitive severe combined immune deficiency [RS-SCID]). Here, we report that Artemis forms a complex with the 469 kDa DNA-dependent protein kinase (DNA-PKcs) in the absence of DNA. The purified Artemis protein alone possesses single-strand-specific 5' to 3' exonuclease activity. Upon complex formation, DNA-PKcs phosphorylates Artemis, and Artemis acquires endonucleolytic activity on 5' and 3' overhangs, as well as hairpins. Finally, the Artemis:DNA-PKcs complex can open hairpins generated by the RAG complex. Thus, DNA-PKcs regulates Artemis by both phosphorylation and complex formation to permit enzymatic activities that are critical for the hairpin-opening step of V(D)J recombination and for the 5' and 3' overhang processing in nonhomologous DNA end joining.

Ku86 is essential in human somatic cells

Ku86 plays a key role in nonhomologous end joining in mammals. Functional inactivation in rodents of either Ku86 or Ku70, which form the heterodimeric DNA end-binding subunit of the DNA-dependent protein kinase complex, is nevertheless compatible with viability. In contrast, no human patient has been described with mutations in either Ku86 or Ku70. This has led to the hypotheses that either these genes are performing an additional essential role(s) and/or redundant pathways exist that mask the phenotypic expression of these genes when they are mutated in humans. To address this issue, we describe here the construction of human somatic cell lines containing a targeted disruption of the Ku86 locus. Human HCT116 colon cancer cells heterozygous for Ku86 were haploinsufficient with an increase in polyploid cells, a reduction in cell proliferation, elevated p53 levels, and a slight hypersensitivity to ionizing radiation. Functional inactivation of the second Ku86 allele resulted in cells with a drastically reduced doubling time. These cells were capable of undergoing only a limited number of cell divisions, after which they underwent apoptosis. These experiments demonstrate that the Ku86 locus is essential in human somatic tissue culture cells.

Highly frequent frameshift DNA synthesis by human DNA polymerase mu

DNA polymerase mu (Polmu) is a newly identified member of the polymerase X family. The biological function of Polmu is not known, although it has been speculated that human Polmu may be a somatic hypermutation polymerase. To help understand the in vivo function of human Polmu, we have performed in vitro biochemical analyses of the purified polymerase. Unlike any other DNA polymerases studied thus far, human Polmu catalyzed frameshift DNA synthesis with an unprecedentedly high frequency. In the sequence contexts examined, -1 deletion occurred as the predominant DNA synthesis mechanism opposite the single-nucleotide repeat sequences AA, GG, TT, and CC in the template. Thus, the fidelity of DNA synthesis by human Polmu was largely dictated by the sequence context. Human Polmu was able to efficiently extend mismatched bases mainly by a frameshift synthesis mechanism. With the primer ends, containing up to four mismatches, examined, human Polmu effectively realigned the primer to achieve annealing with a microhomology region in the template several nucleotides downstream. As a result, human Polmu promoted microhomology search and microhomology pairing between the primer and the template strands of DNA. These results show that human Polmu is much more prone to cause frameshift mutations than base substitutions. The biochemical properties of human Polmu suggest a function in nonhomologous end joining and V(D)J recombination through its microhomology searching and pairing activities but do not support a function in somatic hypermutation.

DNA ligase IV mutations identified in patients exhibiting developmental delay and immunodeficiency

DNA ligase IV functions in DNA nonhomologous end-joining and V(D)J recombination. Four patients with features including immunodeficiency and developmental and growth delay were found to have mutations in the gene encoding DNA ligase IV (LIG4). Their clinical phenotype closely resembles the DNA damage response disorder, Nijmegen breakage syndrome (NBS). Some of the mutations identified in the patients directly disrupt the ligase domain while others impair the interaction between DNA ligase IV and Xrcc-4. Cell lines from the patients show pronounced radiosensitivity. Unlike NBS cell lines, they show normal cell cycle checkpoint responses but impaired DNA double-strand break rejoining. An unexpected V(D)J recombination phenotype is observed involving a small decrease in rejoining frequency coupled with elevated imprecision at signal junctions.

Intermediates in V(D)J recombination: a stable RAG1/2 complex sequesters cleaved RSS ends

Rearrangement of gene segments to generate antigen receptor coding regions depends on the RAG1/2 recombinase, which assembles a synaptic complex between two DNA signal sequences and then cleaves the DNA directly adjacent to the paired signals. After coupled cleavage of complementary signal sequences, virtually all of the cleaved signal ends remained associated with RAG1/2 in stable complexes. These signal end complexes were distinct from various precleavage RAG1/2 signal complexes in that they were resistant to treatment with heparin. A mammalian joining apparatus consisting of purified Ku70/86, XRCC4, and DNA ligase IV proteins was sufficient to join deproteinized cleaved ends, but retention of signal sequences within the signal end complex blocked access to the DNA ends and prevented their joining by these proteins. Sequestration of cleaved ends within the signal end complex would account for the persistence of these ends in the cell after cleavage and may explain why they do not normally activate the DNA-damage-dependent cell cycle checkpoint.

DNA ligase IV-deficient cells are more resistant to ionizing radiation in the absence of Ku70: Implications for DNA double-strand break repair

Vertebrate cells have evolved two major pathways for repairing DNA double-strand breaks (DSBs), homologous recombination (HR) and nonhomologous DNA end-joining (NHEJ). To investigate the role of DNA ligase IV (Lig4) in DSB repair, we knocked out the Lig4 gene (LIG4) in the DT40 chicken B-lymphocyte cell line. The LIG4(-/-) cells showed a marked sensitivity to X-rays, bleomycin, and VP-16 and were more x-ray-sensitive in G(1) than late S or G(2)/M, suggesting a critical role of Lig4 in DSB repair by NHEJ. In support of this notion, HR was not impaired in LIG4(-/-) cells. LIG4(-/-) cells were more x-ray-sensitive when compared with KU70(-/-) DT40 cells, particularly at high doses. Strikingly, however, the x-ray sensitivity of KU70(-/-)/LIG4(-/-) double-mutant cells was essentially the same as that of KU70(-/-) cells, showing that Lig4 deficiency has no effect in the absence of Ku. These results indicate that Lig4 is exclusively required for the Ku-dependent NHEJ pathway of DSB repair and that other DNA ligases (I and III) do not substitute for this function. Our data may explain the observed severe phenotype of Lig4-deficient mice as compared with Ku-deficient mice.

Identification of two topologically independent domains in RAG1 and their role in macromolecular interactions relevant to V(D)J recombination

V(D)J recombination is instigated by the recombination-activating proteins RAG1 and RAG2, which catalyze site-specific DNA cleavage at the border of the recombination signal sequence (RSS). Although both proteins are required for activity, core RAG1 (the catalytically active region containing residues 384-1008 of 1040) alone displays binding specificity for the conserved heptamer and nonamer sequences of the RSS. The nonamer-binding region lies near the N terminus of core RAG1, whereas the heptamer-binding region has not been identified. Here, potential domains within core RAG1 were identified using limited proteolysis studies. An iterative procedure of DNA cloning, protein expression, and characterization revealed the presence of two topologically independent domains within core RAG1, referred to as the central domain (residues 528-760) and the C-terminal domain (residues 761-980). The domains do not include the nonamer-binding region but rather largely span the remaining relatively uncharacterized region of core RAG1. Characterization of macromolecular interactions revealed that the central domain bound to the RSS with specificity for the heptamer and contained the predominant binding site for RAG2. The C-terminal domain bound DNA cooperatively but did not show specificity for either conserved RSS element. This domain was also found to self-associate, implicating it as a dimerization domain within RAG1.

Genomic integrity and the repair of double-strand DNA breaks

The induction of double-strand breaks (DSBs) in DNA by exposure to DNA damaging agents or as intermediates in normal cellular processes, creates a severe threat for the integrity of the genome. Unrepaired or incorrectly repaired DSBs lead to broken chromosomes and/or gross chromosomal rearrangements which are frequently associated with tumor formation in mammals. To maintain the integrity of the genome and to prevent the formation of chromosomal aberrations, several pathways exist in eukaryotes: homologous recombination (HR), non-homologous end joining (NHEJ) and single-strand annealing (SSA). These mechanisms are conserved in evolution, but the relative contribution depends on the organism, cell type and stage of the cell cycle. In yeast, DSBs are primarily repaired via HR while in higher eukaryotes, both HR and NHEJ are important. In mammals, defects in both HR or NHEJ lead to a predisposition to cancer and at the cellular level, the frequency of chromosomal aberrations is increased. This review summarizes our current knowledge about DSB-repair with emphasis on recent progress in understanding the precise biochemical activities of individual proteins involved.

Suppression of gene amplification and chromosomal DNA integration by the DNA mismatch repair system

Mismatch repair (MMR)-deficient cells are shown to produce >15-fold more methotrexate-resistant colonies than MMR normal cells. The increased resistance to methotrexate is primarily due to gene amplification since all the resistant clones contain double-minute chromosomes and increased copy numbers of the DHFR gene. In addition, integration of linearized or retroviral DNAs into chromosomes is also significantly elevated in MMR-deficient cells. These results suggest that in addition to microsatellite instability and homeologous recombination, MMR is also involved in suppression of other genome instabilities such as gene amplification and chromosomal DNA integration.

Homologous DNA recombination in vertebrate cells

The RAD52 epistasis group genes are involved in homologous DNA recombination, and their primary structures are conserved from yeast to humans. Although biochemical studies have suggested that the fundamental mechanism of homologous DNA recombination is conserved from yeast to mammals, recent studies of vertebrate cells deficient in genes of the RAD52 epistasis group reveal that the role of each protein is not necessarily the same as that of the corresponding yeast gene product. This review addresses the roles and mechanisms of homologous recombination-mediated repair with a special emphasis on differences between yeast and vertebrate cells.

Deficiency of human BRCA2 leads to impaired homologous recombination but maintains normal nonhomologous end joining

Carriers of BRCA2 germline mutations are at high risk to develop early-onset breast cancer. The underlying mechanisms of how BRCA2 inactivation predisposes to malignant transformation have not been established. Here, we provide direct functional evidence that human BRCA2 promotes homologous recombination (HR), which comprises one major pathway of DNA double-strand break repair. We found that up-regulated HR after transfection of wild-type (wt) BRCA2 into a human tumor line with mutant BRCA2 was linked to increased radioresistance. In addition, BRCA2-mediated enhancement of HR depended on the interaction with Rad51. In contrast to the tumor suppressor BRCA1, which is involved in multiple DNA repair pathways, BRCA2 status had no impact on the other principal double-strand break repair pathway, nonhomologous end joining. Thus, there exists a specific regulation of HR by BRCA2, which may function to maintain genomic integrity and suppress tumor development in proliferating cells.

Role of the non-homologous DNA end joining pathway in the early steps of retroviral infection

Early after infection, the retroviral RNA genome is reverse transcribed to generate a linear cDNA copy, then that copy is integrated into a chromosome of the host cell. We report that unintegrated viral cDNA is a substrate for the host cell non-homologous DNA end joining (NHEJ) pathway, which normally repairs cellular double-strand breaks by end ligation. NHEJ activity was found to be required for an end-ligation reaction that circularizes a portion of the unintegrated viral cDNA in infected cells. Consistent with this, the NHEJ proteins Ku70 and Ku80 were found to be bound to purified retroviral replication intermediates. Cells defective in NHEJ are known to undergo apoptosis in response to retroviral infection, a response that we show requires reverse transcription to form the cDNA genome but not subsequent integration. We propose that the double-strand ends present in unintegrated cDNA promote apoptosis, as is known to be the case for chromosomal double-strand breaks, and cDNA circularization removes the pro-apoptotic signal.

Cellular and biochemical impact of a mutation in DNA ligase IV conferring clinical radiosensitivity

DNA ligase IV functions in DNA non-homologous end-joining, in V(D)J recombination, and during brain development. We previously reported a homozygous mutation (R278H) in DNA ligase IV in a developmentally normal leukemia patient who overresponded to radiotherapy. The impact of this hypomorphic mutation has been evaluated using cellular, biochemical, and structural approaches. Structural modeling using T7 DNA ligase predicts that the activity and conformational stability of the protein is likely to be impaired. We show that wild type DNA ligase IV-Xrcc4 is an efficient double-stranded ligase with distinct optimal requirements for adenylate complex formation versus rejoining. The mutation impairs the formation of an adenylate complex as well as reducing the rejoining activity. Additionally, it imparts temperature-sensitive activity to the protein consistent with the predictions of the structural modeling. At the cellular level, the mutation confers a unique V(D)J recombination phenotype affecting the fidelity of signal joint formation with little effect on the frequency of the reaction. These findings suggest that hypomorphic mutations in ligase IV may allow normal development but confer marked radiosensitivity.

Efficient rejoining of radiation-induced DNA double-strand breaks in vertebrate cells deficient in genes of the RAD52 epistasis group

Rejoining of ionizing radiation (IR) induced DNA DSBs usually follows biphasic kinetics with a fast (t(50): 5-30 min) component attributed to DNA-PK-dependent non-homologous endjoining (NHEJ) and a slow (t(50): 1-20 h), as of yet uncharacterized, component. To examine whether homologous recombination (HR) contributes to DNA DSB rejoining, a systematic genetic study was undertaken using the hyper-recombinogenic DT40 chicken cell line and a series of mutants defective in HR. We show that DT40 cells rejoin IR-induced DNA DSBs with half times of 13 min and 4.5 h and contributions by the fast (78%) and the slow (22%) components similar to those of other vertebrate cells with 1000-fold lower levels of HR. We also show that deletion of RAD51B, RAD52 and RAD54 leaves unchanged the rejoining half times and the contribution of the slow component, as does also a conditional knock out mutant of RAD51. A significant reduction (to 37%) in the contribution of the fast component is observed in Ku70(-/-) DT40 cells, but the slow component, operating with a half time of 18.4 h, is still able to rejoin the majority (63%) of DSBs. A double mutant Ku70(-/-)/RAD54(-/-) shows similar half times to Ku70(-/-) cells. Thus, variations in HR by several orders of magnitude leave unchanged the kinetics of rejoining of DNA DSBs, and fail to modify the contribution of the slow component in a way compatible with a dependence on HR. We propose that, in contrast to yeast, cells of vertebrates are 'hard-wired' in the utilization of NHEJ as the main pathway for rejoining of IR-induced DNA DSBs and speculate that the contribution of homologous recombination repair (HRR) is at a stage after the initial rejoining.

Genetic evidence for the involvement of DNA ligase IV in the DNA-PK-dependent pathway of non-homologous end joining in mammalian cells

Cells of vertebrates remove DNA double-strand breaks (DSBs) from their genome predominantly utilizing a fast, DNA-PKcs-dependent form of non-homologous end joining (D-NHEJ). Mutants with inactive DNA-PKcs remove the majority of DNA DSBs utilizing a slow, DNA-PKcs-independent pathway that does not utilize genes of the RAD52 epistasis group, is error-prone and can therefore be classified as a form of NHEJ (termed basic or B-NHEJ). We studied the role of DNA ligase IV in these pathways of NHEJ. Although biochemical studies show physical and functional interactions between the DNA-PKcs/Ku and the DNA ligase IV/Xrcc4 complexes suggesting operation within the same pathway, genetic evidence to support this notion is lacking in mammalian cells. Primary human fibroblasts (180BR) with an inactivating mutation in DNA ligase IV, rejoined DNA DSBs predominantly with slow kinetics similar to those observed in cells deficient in DNA-PKcs, or in wild-type cells treated with wortmannin to inactivate DNA-PK. Treatment of 180BR cells with wortmannin had only a small effect on DNA DSB rejoining and no effect on cell radiosensitivity to killing although it sensitized control cells to 180BR levels. This is consistent with DNA ligase IV functioning as a component of the D-NHEJ, and demonstrates the unperturbed operation of the DNA-PKcs-independent pathway (B-NHEJ) at significantly reduced levels of DNA ligase IV. In vitro, extracts of 180BR cells supported end joining of restriction endonuclease-digested plasmid to the same degree as extracts of control cells when tested at 10 mM Mg(2+). At 0.5 mM Mg(2+), where only DNA ligase IV is expected to retain activity, low levels of end joining ( approximately 10% of 10 mM) were seen in the control but there was no detectable activity in 180BR cells. Antibodies raised against DNA ligase IV did not measurably inhibit end joining at 10 mM Mg(2+) in either cell line. Thus, in contrast to the situation in vivo, end joining in vitro is dominated by pathways with properties similar to B-NHEJ that do not display a strong dependence on DNA ligase IV, with D-NHEJ retaining only a limited contribution. The implications of these observations to studies of NHEJ in vivo and in vitro are discussed.

How to make ends meet in V(D)J recombination

In most vertebrate species analyzed so far, the diversity of soluble or membrane-bound antigen-receptors expressed by B and T lymphocytes is generated by V(D)J recombination. During this process, the coding regions for the variable domains of antigen-receptors are created by the joining of subexons that are randomly selected from arrays of tandemly repeated V, D (sometimes) and J gene segments. This involves the site-specific cleavage of chromosomal DNA by the lymphocyte-specific recombination-activating gene (RAG)-1/2 proteins, which appear to have originated from an ancient transposable element. The DNA double-strand breaks created by RAG proteins are subsequently processed and rejoined by components of the nonhomologous DNA end-joining pathway, which is conserved in all eukaryotic organisms - from unicellular yeast up to highly complex mammalian species.

Identification of a mutated DNA ligase IV gene in the X-ray-hypersensitive mutant SX10 of mouse FM3A cells

The mouse carcinoma cell line SX10 is a hypersensitive mutant to x-rays and bleomycin. An earlier complementation test suggests that SX10 would belong to x-ray-cross complementing group (XRCC) 4. However, in this study, a human XRCC4 expression vector failed to complement the SX10 phenotype. Consistent with the previous report, SX10 showed the same level of DNA-dependent protein kinase activity as the wild-type SR-1. We isolated and analyzed hybrids between SX10 and human diploid fibroblast cells and found that human chromosome 13 conferred the x-ray resistance to the hybrids, suggesting that a candidate gene would be located on this chromosome. Polymerase chain reaction analysis with these hybrids and x-ray-resistant transformants obtained by introducing human chromosomes into SX10 indicated that the mutant was likely to be defective in DNA ligase IV. Sequence analysis of the DNA ligase IV gene confirmed that a defect in SX10 was attributed to a transition of G to A at nucleotide position 1413 of the gene, leading to an amino acid substitution from Trp at residue 471 to a stop codon. Revertant clones (Rev1-3) derived from SX10 showed a restored x-ray resistance; Rev1 reverted to the original nucleotide G at position 1413, whereas Rev2 and Rev3 to C. Transfection of a mouse DNA ligase IV cDNA vector into SX10 restored the resistance to both x-rays and bleomycin. SX10 showed a reduced frequency of chromosomal integration of transfected DNA, but the revertants restored the frequency found in the wild-type cells. These results suggest a possible involvement of DNA ligase IV in the integration event of foreign DNA as well as a crucial role in DNA double-strand break repair.

Chromosomal stability and the DNA double-stranded break connection

Genome stability is of primary importance for the survival and proper functioning of all organisms. Double-stranded breaks in DNA are important threats to genome integrity because they can result in chromosomal aberrations that can affect, simultaneously, many genes, and lead to cell malfunctioning and cell death. These detrimental consequences are counteracted by two mechanistically distinct pathways of double-stranded break repair: homologous recombination and non-homologous end-joining. Recently, unexpected links between these double-stranded break-repair systems, and several human genome instability and cancer predisposition syndromes, have emerged. Now, interactions between both double-stranded break-repair pathways and other cellular processes, such as cell-cycle regulation and replication, are being unveiled.

Linear DNAs concatemerize in vivo and result in sustained transgene expression in mouse liver

The short duration of transgene expression remains a major obstacle for the implementation of nonviral DNA vectors in clinical gene therapy trials. Here, we demonstrate stable, long-term transgene expression in vivo by transfecting a linear DNA expression cassette (LDNA) into mouse liver. Interestingly, despite similar quantities and cellular distribution of injected DNAs in their livers, mice receiving LDNA encoding human alpha1-antitrypsin (hAAT) expressed approximately 10- to 100-fold more serum hAAT than mice injected with closed circular (cc) DNA for a period of 9 months (length of study). Furthermore, when a linear human factor IX expression cassette was delivered to factor IX-deficient mice, sustained serum concentrations of more than 4 microg/ml (80% of normal) of the human clotting factor and correction of the bleeding diathesis were obtained. Southern blot analyses indicate that, unlike ccDNA, LDNA rapidly formed large, unintegrated concatemers in vivo, suggesting that transgene persistence from plasmid-based vectors was influenced by the structure of the vector in transfected cells. No differences in transgene expression or DNA molecular structures were observed when AAV ITRs were included to flank the hAAT expression cassette in both ccDNA- and LDNA-treated animals. Linear DNA transfection provides an approach for achieving long-term expression of a transgene in vivo.

Antisense-mediated decrease in DNA ligase III expression results in reduced mitochondrial DNA integrity

The human DNA ligase III gene encodes both nuclear and mitochondrial proteins. Abundant evidence supports the conclusion that the nuclear DNA ligase III protein plays an essential role in both base excision repair and homologous recombination. However, the role of DNA ligase III protein in mitochondrial genome dynamics has been obscure. Human tumor-derived HT1080 cells were transfected with an antisense DNA ligase III expression vector and clones with diminished levels of DNA ligase III activity identified. Mitochondrial protein extracts prepared from these clones had decreased levels of DNA ligase III relative to extracts from cells transfected with a control vector. Analysis of these clones revealed that the DNA ligase III antisense mRNA-expressing cells had reduced mtDNA content compared to control cells. In addition, the residual mtDNA present in these cells had numerous single-strand nicks that were not detected in mtDNA from control cells. Cells expressing antisense ligase III also had diminished capacity to restore their mtDNA to pre-irradiation levels following exposure to gamma-irradiation. An antisense-mediated reduction in cellular DNA ligase IV had no effect on the copy number or integrity of mtDNA. This observation, coupled with other evidence, suggests that DNA ligase IV is not present in the mitochondria and does not play a role in maintaining mtDNA integrity. We conclude that DNA ligase III is essential for the proper maintenance of mtDNA in cultured mammalian somatic cells.

Human DNA repair genes

DNA repair systems are essential for the maintenance of genome integrity. Consequently, the disregulation of repair genes can be expected to be associated with significant, detrimental health effects, which can include an increased prevalence of birth defects, an enhancement of cancer risk, and an accelerated rate of aging. Although original insights into DNA repair and the genes responsible were largely derived from studies in bacteria and yeast, well over 125 genes directly involved in DNA repair have now been identified in humans, and their cDNA sequence established. These genes function in a diverse set of pathways that involve the recognition and removal of DNA lesions, tolerance to DNA damage, and protection from errors of incorporation made during DNA replication or DNA repair. Additional genes indirectly affect DNA repair, by regulating the cell cycle, ostensibly to provide an opportunity for repair or to direct the cell to apoptosis. For about 70 of the DNA repair genes listed in Table I, both the genomic DNA sequence and the cDNA sequence and chromosomal location have been elucidated. In 45 cases single-nucleotide polymorphisms have been identified and, in some cases, genetic variants have been associated with specific disorders. With the accelerating rate of gene discovery, the number of identified DNA repair genes and sequence variants is quickly rising. This report tabulates the current status of what is known about these genes. The report is limited to genes whose function is directly related to DNA repair.

cDNA expression array analysis of DNA repair genes in human glioma cells that lack or express DNA-PK

M059J cells provide the only example of DNA-PKcs (now known as PRKDC) deficiency in a human cell line. M059K cells, derived from the same tumor specimen, express PRKDC protein and activity and, together with M059J, provide a useful model in which to study the role of DNA-PK in cellular responses to DNA-damaging agents. Because these cells are of tumor origin, we used Atlas human cancer cDNA expression arrays to investigate possible differential expression of other DNA repair genes in control and irradiated samples. cDNA array results indicated differential expression of 14 genes. Northern blotting confirmed relatively greater expression of replication factor C 37-kDa subunit mRNA in M059J cells compared to M059K cells and reduced expression of DNA ligase IV compared to ligase III in both cell lines independent of irradiation. These results suggest that other DNA repair proteins are altered in these cell lines and that repair mechanisms predicted from the study of normal tissues may be fundamentally altered in human cancer cells.

The scid recombination-inducible cell line: a model to study DNA-PK-independent V(D)J recombination

To investigate the molecular mechanisms of the variable (diversity) joining (V(D)J) recombination process at an endogenous gene locus, recombination-inducible cell lines were made from both bcl-2-bearing severe combined immune deficiency (scid) homozygous and scid heterozyous (s/ + ) mice by transforming pre-B cells with the temperature-sensitive Abelson murine leukemia virus (ts-Ab-MLV). These transformants can be induced to undergo immunoglobulin light-chain gene rearrangements by incubating them at the non-permissive temperature. In the case of transformed scid cells, a significant amount of hairpin coding ends are accumulated during recombination induction, but few coding joints are generated. After being shifted to the permissive temperature. however, these cells are capable of opening hairpin ends and forming coding joints. Thus, ts-Ab-MLV transformed scid cells can be readily manipulated for both recombination cleavage and end resolution. However, unlike the rapid coding joint formation in s/ + cells that have the catalytic subunit of DNA-dependent protein kinase (DNA-PKcs), the process for resolving coding ends in scid cells is slow and error prone, and also appears to be correlated with a reduction in the RAG1/2 expression. Apparently, this process is mediated by a DNA-PK-independent pathway. The fact that the activity of this pathway can be manipulated in vitro makes it possible to delineate the mechanisms in end opening, processing and joining. Therefore, these ts-Ab-MLV transformed scid cell lines offer a model to study the molecular nature as well as the regulation of the DNA-PK-independent pathway in coding end resolution.

Coupled homologous and nonhomologous repair of a double-strand break preserves genomic integrity in mammalian cells

DNA double-strand breaks (DSBs) may be caused by normal metabolic processes or exogenous DNA damaging agents and can promote chromosomal rearrangements, including translocations, deletions, or chromosome loss. In mammalian cells, both homologous recombination and nonhomologous end joining (NHEJ) are important DSB repair pathways for the maintenance of genomic stability. Using a mouse embryonic stem cell system, we previously demonstrated that a DSB in one chromosome can be repaired by recombination with a homologous sequence on a heterologous chromosome, without any evidence of genome rearrangements (C. Richardson, M. E. Moynahan, and M. Jasin, Genes Dev., 12:3831-3842, 1998). To determine if genomic integrity would be compromised if homology were constrained, we have now examined interchromosomal recombination between truncated but overlapping gene sequences. Despite these constraints, recombinants were readily recovered when a DSB was introduced into one of the sequences. The overwhelming majority of recombinants showed no evidence of chromosomal rearrangements. Instead, events were initiated by homologous invasion of one chromosome end and completed by NHEJ to the other chromosome end, which remained highly preserved throughout the process. Thus, genomic integrity was maintained by a coupling of homologous and nonhomologous repair pathways. Interestingly, the recombination frequency, although not the structure of the recombinant repair products, was sensitive to the relative orientation of the gene sequences on the interacting chromosomes.

DNA ligase IV and XRCC4 form a stable mixed tetramer that functions synergistically with other repair factors in a cell-free end-joining system

Repair of DNA double-strand breaks in mammalian cells occurs via a direct nonhomologous end-joining pathway. Although this pathway can be studied in vivo and in crude cell-free systems, a deeper understanding of the mechanism requires reconstitution with purified enzymes. We have expressed and purified a complex of two proteins that are critical for double-strand break repair, DNA ligase IV (DNL IV) and XRCC4. The complex is homogeneous, with a molecular mass of about 300,000 Da, suggestive of a mixed tetramer containing two copies of each polypeptide. The presence of multiple copies of DNL IV was confirmed in an experiment where different epitope-tagged forms of DNL IV were recovered simultaneously in the same complex. Cross-linking suggests that an XRCC4.XRCC4 dimer interface forms the core of the tetramer, and that the DNL IV polypeptides are in contact with XRCC4 but not with one another. Purified DNL IV.XRCC4 complex functioned synergistically with Ku protein, the DNA-dependent protein kinase catalytic subunit, and other repair factors in a cell-free end-joining assay. We suggest that a dyad-symmetric DNL IV.XRCC4 tetramer bridges the two ends of the broken DNA and catalyzes the coordinate ligation of the two DNA strands.

Normal V(D)J recombination in cells from patients with Nijmegen breakage syndrome

The majority of antigen receptor diversity in mammals is generated by V(D)J recombination. During this process DNA double strand breaks are introduced at recombination signals by lymphoid specific RAG1/2 proteins generating blunt ended signal ends and hairpinned coding ends. Rejoining of all DNA ends requires ubiquitously expressed DNA repair proteins, such as Ku70/86 and DNA ligase IV/XRCC4. In addition, the formation of coding joints depends on the function of the scid gene encoding the catalytic subunit of DNA-dependent protein kinase, DNA-PK(CS), that is somehow required for processing of coding end hairpins. Recently, it was shown that purified RAG1/2 proteins can cleave DNA hairpins in vitro, but the same activity was also described for a protein complex of the DNA repair proteins Nbs1/Mre11/Rad50. This leaves the possibility that either protein complex might be involved in coding end processing in V(D)J recombination. We have therefore analyzed V(D)J recombination in cells from patients with Nijmegen breakage syndrome, carrying a mutation in the nbs1 gene. We find that V(D)J recombination frequencies and the quality of signal and coding joining are comparable to wild-type controls, as analyzed by a cellular V(D)J recombination assay. In addition, we did not detect significant differences in CDR3 sequences of endogenous Ig lambdaL and kappaL chain gene loci cloned from peripheral blood lymphocytes of an NBS patient and of healthy individuals. These findings suggest that the Nbs1/Mre11/Rad50 complex is not involved in coding end processing of V(D)J recombination.

Telomere dysfunction impairs DNA repair and enhances sensitivity to ionizing radiation

Telomeres are specialized nucleoprotein complexes that serve as protective caps of linear eukaryotic chromosomes. Loss of telomere function is associated with rampant genetic instability and loss of cellular viability and renewal potential. The telomere also participates in processes of chromosomal repair, as evidenced by the 'capture' or de novo synthesis of telomere repeats at double-stranded breaks and by the capacity of yeast telomeres to serve as repositories of essential components of the DNA repair machinery, particularly those involved in non-homologous end-joining (NHEJ). Here we used the telomerase-deficient mouse, null for the essential telomerase RNA gene (Terc), to assess the role of telomerase and telomere function on the cellular and organismal response to ionizing radiation. Although the loss of telomerase activity per se had no discernable impact on the response to ionizing radiation, the emergence of telomere dysfunction in late-generation Terc-/- mice imparted a radiosensitivity syndrome associated with accelerated mortality. On the cellular level, the gastrointestinal crypt stem cells and primary thymocytes showed increased rates of apoptosis, and mouse embryonic fibroblasts (MEFs) showed diminished dose-dependent clonogenic survival. The radiosensitivity of telomere dysfunctional cells correlated with delayed DNA break repair kinetics, persistent chromosomal breaks and cytogenetic profiles characterized by complex chromosomal aberrations and massive fragmentation. Our findings establish a intimate relationship between functionally intact telomeres and the genomic, cellular and organismal response to ionizing radiation.

Interactions of the DNA ligase IV-XRCC4 complex with DNA ends and the DNA-dependent protein kinase

The DNA-dependent protein kinase (DNA-PK), consisting of Ku and the DNA-PK catalytic subunit (DNA-PKcs), and the DNA ligase IV-XRCC4 complex function together in the repair of DNA double-strand breaks by non-homologous end joining. These protein complexes are also required for the completion of V(D)J recombination events in immune cells. Here we demonstrate that the DNA ligase IV-XRCC4 complex binds specifically to the ends of duplex DNA molecules and can act as a bridging factor, linking together duplex DNA molecules with complementary but non-ligatable ends. Although the DNA end-binding protein Ku inhibited DNA joining by DNA ligase IV-XRCC4, it did not prevent this complex from binding to DNA. Instead, DNA ligase IV-XRCC4 and Ku bound simultaneously to the ends of duplex DNA molecules. DNA ligase IV-XRCC4 and DNA-PKcs also formed complexes at the ends of DNA molecules, but DNA-PKcs did not inhibit ligation. Interestingly, DNA-PKcs stimulated intermolecular ligation by DNA ligase IV-XRCC4. In the presence of DNA-PK, the majority of the joining events catalyzed by DNA ligase IV-XRCC4 were intermolecular because Ku inhibited intramolecular ligation, but DNA-PKcs still stimulated intramolecular ligation. We suggest that DNA-PKcs-containing complexes formed at DNA ends enhance the association of DNA ends via protein-protein interactions, thereby stimulating intermolecular ligation.

Both V(D)J recombination and radioresistance require DNA-PK kinase activity, though minimal levels suffice for V(D)J recombination

DNA-dependent protein kinase (DNA-PK) is utilized in both DNA double-strand break repair (DSBR) and V(D)J recombination, but the mechanism by which this multiprotein complex participates in these processes is unknown. To evaluate the importance of DNA-PK-mediated protein phosphorylation in DSBR and V(D)J recombination, we assessed the effects of the phosphatidyl inositol 3-kinase inhibitor wortmannin on the repair of ionizing radiation-induced DNA double-strand breaks and V(D)J recombination in the V(D)J recombinase inducible B cell line HDR37. Wortmannin radiosensitized HDR37, but had no affect on V(D)J recombination despite a marked reduction in DNA-PK activity. On the other hand, studies with mammalian expression vectors for wild-type human DNA-PK catalytic subunit (DNA-PKcs) and a kinase domain mutant demonstrated that only the kinase active form of DNA-PKcs can reconstitute DSBR and V(D)J recombination in a DNA-PKcs-deficient cell line (Sf19), implying that DNA-PKcs kinase activity is essential for both DSBR and V(D)J recombination. These apparently contradictory results were reconciled by analyses of cell lines varying in their expression of recombinant wild-type human DNA-PKcs. These studies establish that minimal DNA-PKcs protein levels are sufficient to support V(D)J recombination, but insufficient to confer resistance to ionizing radiation.

Tying up loose ends: nonhomologous end-joining in Saccharomyces cerevisiae

The ends of chromosomal DNA double-strand breaks (DSBs) can be accurately rejoined by at least two discrete pathways, homologous recombination and nonhomologous end-joining (NHEJ). The NHEJ pathway is essential for repair of specific classes of DSB termini in cells of the budding yeast Saccharomyces cerevisiae. Endonuclease-induced DSBs retaining complementary single-stranded DNA overhangs are repaired efficiently by end-joining. In contrast, damaged DSB ends (e.g., termini produced by ionizing radiation) are poor substrates for this pathway. NHEJ repair involves the functions of at least 10 genes, including YKU70, YKU80, DNL4, LIF1, SIR2, SIR3, SIR4, RAD50, MRE11, and XRS2. Most or all of these genes are required for efficient recombination-independent recircularization of linearized plasmids and for rejoining of EcoRI endonuclease-induced chromosomal DSBs in vivo. Several NHEJ mutants also display aberrant processing and rejoining of DSBs that are generated by HO endonuclease or formed spontaneously in dicentric plasmids. In addition, all NHEJ genes except DNL4 and LIF1 are required for stabilization of telomeric repeat sequences. Each of the proteins involved in NHEJ appears to bind, directly or through protein associations, with the ends of linear DNA. Enzymatic and/or structural roles in the rejoining of DSB termini have been postulated for several proteins within the group. Most yeast NHEJ genes have homologues in human cells and many biochemical activities and protein:protein interactions have been conserved in higher eucaryotes. Similarities and differences between NHEJ repair in yeast and mammalian cells are discussed.

Evidence that terminal deoxynucleotidyltransferase expression plays a role in Ig heavy chain gene segment utilization

TdT is a nuclear enzyme that catalyzes the addition of random nucleotides at Ig and TCR V(D)J junctions. In this paper we analyze human IgH rearrangements generated from transgenic minilocus mice in the presence or absence of TdT. In the absence of TdT, the pseudo-VH gene segment present in the minilocus is rearranged dramatically more frequently. Additionally, JH6 gene segment utilization is increased as well as the number of rearrangements involving only VH and JH gene segments. Thus, the recombination of IgH gene segments that are flanked by 23-nt spacer recombination signal sequences may be influenced by TdT expression. Extensive analysis indicates that these changes are independent of antigenic selection and cannot be explained by homology-mediated recombination. Thus, the role played by TdT may be more extensive than previously thought.

Ku recruits the XRCC4-ligase IV complex to DNA ends

Genetic experiments have determined that Ku, XRCC4, and ligase IV are required for repair of double-strand breaks by the end-joining pathway. The last two factors form a tight complex in cells. However, ligase IV is only one of three known mammalian ligases and is intrinsically the least active in intermolecular ligation; thus, the biochemical basis for requiring this ligase has been unclear. We demonstrate here a direct physical interaction between the XRCC4-ligase IV complex and Ku. This interaction is stimulated once Ku binds to DNA ends. Since XRCC4-ligase IV alone has very low DNA binding activity, Ku is required for effective recruitment of this ligase to DNA ends. We further show that this recruitment is critical for efficient end-joining activity in vitro. Preformation of a complex containing Ku and XRCC4-ligase IV increases the initial ligation rate 20-fold, indicating that recruitment of the ligase is an important limiting step in intermolecular ligation. Recruitment by Ku also allows XRCC4-ligase IV to use Ku's high affinity for DNA ends to rapidly locate and ligate ends in an excess of unbroken DNA, a necessity for end joining in cells. These properties are conferred only on ligase IV, because Ku does not similarly interact with the other mammalian ligases. We have therefore defined cell-free conditions that reflect the genetic requirement for ligase IV in cellular end joining and consequently can explain in molecular terms why this factor is required.

Metabolism of recombination coding ends in scid cells

V(D)J recombination cleavage generates two types of dsDNA breaks: blunt signal ends and covalently sealed hairpin coding ends. Although signal ends can be directly ligated to form signal joints, hairpin coding ends need to be opened and subsequently processed before being joined. However, the underlying mechanism of coding end resolution remains undefined. The current study attempts to delineate this process by analyzing various structures of coding ends made in situ from recombination-inducible pre-B cell lines of both normal and scid mice. These cell lines were derived by transformation of B cell precursors with the temperature-sensitive Abelson murine leukemia virus. Our kinetic analysis revealed that under conditions permissive to scid transformants, hairpin coding ends could be nicked to generate 3' overhangs and then processed into blunt ends. The final joining of these blunt ends followed the same kinetics as signal joint formation. The course of this process is in sharp contrast to coding end resolution in scid heterozygous transformants that express the catalytic subunit of DNA-dependent protein kinase, in which hairpin end opening, processing, and joining proceeded very rapidly and appeared to be closely linked. Furthermore, we demonstrated that the opening of hairpin ends in scid cells could be manipulated by different culture conditions, which ultimately influenced not only the level and integrity of the newly formed coding joints, but also the extent of microhomology at the coding junctions. These results are discussed in the context of scid leaky recombination.

Neonatal lethality with abnormal neurogenesis in mice deficient in DNA polymerase beta

DNA polymerase beta (Polbeta) has been implicated in base excision repair in mammalian cells. However, the physiological significance of this enzyme in the body remains unclear. Here, we demonstrate that mice carrying a targeted disruption of the Polbeta gene showed growth retardation and died of a respiratory failure immediately after the birth. Histological examination of the embryos revealed defective neurogenesis characterized by apoptotic cell death in the developing central and peripheral nervous systems. Extensive cell death occurred in newly generated post-mitotic neuronal cells and was closely associated with the period between onset and cessation of neurogenesis. These findings indicate that Polbeta plays an essential role in neural development.

Defective embryonic neurogenesis in Ku-deficient but not DNA-dependent protein kinase catalytic subunit-deficient mice

Mammalian nonhomologous DNA end joining employs Ku70, Ku80, DNA-dependent protein kinase catalytic subunit (DNA-PKcs), XRCC4, and DNA ligase IV (Lig4). Herein, we show that Ku70 and Ku80 deficiency but not DNA-PKcs deficiency results in dramatically increased death of developing embryonic neurons in mice. The Ku-deficient phenotype is qualitatively similar to, but less severe than, that associated with XRCC4 and Lig4 deficiency. The lack of a neuronal death phenotype in DNA-PKcs-deficient embryos and the milder phenotype of Ku-deficient versus XRCC4- or Lig4-deficient embryos correlate with relative leakiness of residual end joining in these mutant backgrounds as assayed by a V(D)J recombination end joining assay. We conclude that normal development of the nervous system depends on the four evolutionarily conserved nonhomologous DNA end joining factors.

Analyses of TCRB rearrangements substantiate a profound deficit in recombination signal sequence joining in SCID foals: implications for the role of DNA-dependent protein kinase in V(D)J recombination

We reported previously that the genetic SCID disease observed in Arabian foals is explained by a defect in V(D)J recombination that profoundly affects both coding and signal end joining. As in C.B-17 SCID mice, the molecular defect in SCID foals is in the catalytic subunit of the DNA-dependent protein kinase (DNA-PKCS); however, in SCID mice, signal end resolution remains relatively intact. Moreover, recent reports indicate that mice that completely lack DNA-PKCS also generate signal joints at levels that are indistinguishable from those observed in C.B-17 SCID mice, eliminating the possibility that a partially active version of DNA-PKCS facilitates signal end resolution in SCID mice. We have analyzed TCRB rearrangements and find that signal joints are reduced by approximately 4 logs in equine SCID thymocytes as compared with normal horse thymocytes. A potential explanation for the differences between SCID mice and foals is that the mutant DNA-PKCS allele in SCID foals inhibits signal end resolution. We tested this hypothesis using DNA-PKCS expression vectors; in sum, we find no evidence of a dominant-negative effect by the mutant protein. These and other recent data are consistent with an emerging consensus: that in normal cells, DNA-PKCS participates in both coding and signal end resolution, but in the absence of DNA-PKCS an undefined end joining pathway (which is variably expressed in different species and cell types) can facilitate imperfect signal and coding end joining.

Transposition mediated by RAG1 and RAG2 and the evolution of the adaptive immune system

The RAG1 and RAG2 proteins together initiate V(D)J recombination by performing cleavage of chromosomal DNA adjacent to antigen receptor gene segments. Like the adaptive immune system itself, RAG1 and RAG2 are found only in jawed vertebrates. The hypothesis that RAG1 and RAG2 arose in evolution as components of a transposable element has received dramatic support from our recent finding that the RAG proteins are a fully functional transposase in vitro. This result strongly suggests that antigen receptor genes acquired their unusual structure as a consequence of the insertion of a transposable element into an ancestral receptor gene by RAG1 and RAG2 approx 450 million years ago.

Coordinate regulation of RAG1 and RAG2 by cell type-specific DNA elements 5' of RAG2

RAG1 and RAG2 are essential for V(D)J recombination and lymphocyte development. These genes are thought to encode a transposase derived from a mobile genetic element that was inserted into the vertebrate genome 450 million years ago. The regulation of RAG1 and RAG2 was investigated in vivo with bacterial artificial chromosome (BAC) transgenes containing a fluorescent indicator. Coordinate expression of RAG1 and RAG2 in B and T cells was found to be regulated by distinct genetic elements found on the 5' side of the RAG2 gene. This observation suggests a mechanism by which asymmetrically disposed cis DNA elements could influence the expression of the primordial transposon and thereby capture RAGs for vertebrate evolution.