SEARCH

SEARCH BY CITATION

Keywords:

  • DNA repair;
  • non-homologous end joining;
  • BRCT domain

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Rejoining of single- and double-strand breaks (DSBs) introduced in DNA during replication, recombination, and DNA damage is catalysed by DNA ligase enzymes. Eukaryotes possess multiple DNA ligase enzymes, each having distinct roles in cellular metabolism. Double-strand breaks in DNA, which can occur spontaneously in the cell or be induced experimentally by γ-irradiation, represent one of the most serious threats to genomic integrity. Non-homologous end joining (NHEJ) rather than homologous recombination is the major pathway for repair of DSBs in organisms with complex genomes, including humans and plants. DNA ligase IV in Saccharomyces cerevisiae and humans catalyses the final step in the NHEJ pathway of DSB repair. In this study we identify an Arabidopsis thaliana homologue (AtLIG4) of human and S. cerevisiae DNA ligase IV which is shown to encode an ATP-dependent DNA ligase with a theoretical molecular mass of 138 kDa and 48% similarity in amino-acid sequence to the human DNA ligase IV. Yeast two-hybrid analysis demonstrated a strong interaction between A. thaliana DNA ligase IV and the A. thaliana homologue of the human DNA ligase IV-binding protein XRCC4. This interaction is shown to be mediated via the tandem BRCA C-terminal domains of A. thaliana DNA ligase IV protein. Expression of AtLIG4 is induced by γ-irradiation but not by UVB irradiation, consistent with an in vivo role for the A. thaliana DNA ligase IV in DSB repair.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Discontinuities in the DNA backbone arise during DNA replication, recombination and excision repair, and can also be caused by a range of DNA-damaging agents ( Wood, 1996). These single- and double-strand breaks (DSBs) in the phosphodiester backbone of DNA are rejoined by the action of DNA ligases. In prokaryotes, one NAD-dependent DNA ligase enzyme catalyses the final ligation step in replication, recombination and repair pathways. However, eukaryotes possess multiple forms of ATP-dependent DNA ligases, each of which has specific roles in cellular DNA metabolism (reviewed in Tomkinson and Mackey, 1998). In humans (Homo sapiens) and Saccharomyces cerevisiae, DNA ligase I is responsible both for joining Okazaki fragments during DNA replication and also ligating single-strand nicks formed during excision repair ( Barnes et al., 1992 ; Li et al., 1994 ). A DNA ligase I homologue has been cloned from Arabidopsis thaliana and shown to complement a conditional S. cerevisiae DNA ligase I mutant ( Taylor et al., 1998a ). Expression of A. thaliana DNA ligase I was found to be highest in rapidly dividing meristematic tissue, suggesting a role for this ligase in plants similar to that of DNA ligase I in humans and S. cerevisiae. Biochemical analysis of human cell extracts identified three further DNA ligase activities, designated DNA ligase II–IV. Human DNA ligase III is expressed as two splice variants. DNA ligase IIIα joins single-strand breaks in DNA; is associated with base excision repair; and binds to XRCC1, a protein required for the repair of single-strand breaks induced by γ-irradiation and alkylating reagents ( Caldecott et al., 1994 ). DNA ligase IIIβ is expressed only in the testis and may have a role in meiotic recombination ( Mackey et al., 1996 ). Human DNA ligase III also encodes a mitochondrial form, probably produced by translation initiation at an upstream translation start site, producing a protein with a mitochondrial targeting sequence ( Lakshmipathy and Campbell, 1999). Human DNA ligase II, a 70 kDa protein when purified from liver, is believed to be derived from partial proteolysis of DNA ligase III. Analysis of the complete S. cerevisiae genome sequence for the presence of the highly conserved DNA ligase catalytic domain consensus sequence found no S. cerevisiae ligase III homologue ( Teo and Jackson, 1997). This result raised the question of the genetic basis of mitochondrial DNA ligase activity in S. cerevisiae. This has been answered recently with the finding that S. cerevisiae DNA ligase I has alternative translation start sites, producing DNA ligase protein with or without a mitochondrial targeting presequence ( Willer et al., 1999 ).

Analysis of the S. cerevisiae genomic sequence did identify an S. cerevisiae homologue of human DNA ligase IV involved in ligation of DSBs, which are one of the most serious threats to the integrity of genomic DNA ( Teo and Jackson, 1997). In plants, as in other higher eukaryotes, DSBs are repaired mainly by direct end-to-end joining of DNA strands independent of sequence homology. This process of illegitimate recombination is called non-homologous end joining (NHEJ) and, unlike homologous recombination, can often produce deletions and genetic rearrangements. NHEJ is often associated with deletions and insertions of random filler DNA corresponding to genomic DNA and/or plasmid or T-DNA, which have been found in plasmids recircularized in tobacco protoplasts and in the mutational repair of a highly specific I-SceI restriction site in a transgene in the tobacco genome ( Gorbunova and Levy, 1997; Gorbunova and Levy, 1999; Salomon and Puchta, 1998). The fact that NHEJ is the main pathway of DSB repair has proved an obstacle to the development of gene targeting and gene replacement technologies in plants. Attempts to promote homologous recombination in higher plants by flanking the transgene with long stretches of genomic DNA have been unsuccessful, and the ratio of homologous recombination to NHEJ remains low, of the order 1 : 10−3−10−6 ( Liljegren and Yanofsky, 1998; Mengiste and Paszkowski, 1999; Puchta, 1998).

Recent advances have identified some of the molecular components involved in NHEJ, and have suggested a mechanism for NHEJ in S. cerevisiae and mammals ( Chu, 1997; Critchlow and Jackson, 1998; Kanaar et al., 1998 ). The ligation step is catalysed by DNA ligase IV, and in S. cerevisiae a mutation of DNA ligase IV severely impairs end joining. This observation suggests that there is little overlap in the substrate specificity between S. cerevisiae DNA ligases I and IV in vivo ( Schär et al., 1997 ; Teo and Jackson, 1997; Wilson et al., 1997 ). In A. thaliana there is also evidence against overlapping functions of the DNA ligases. A mutation in DNA ligase I cannot be complemented by DNA ligase IV, as the mutation in DNA ligase I is lethal in homozygous mutant plants ( Babiychuk et al., 1998 ). Human and S. cerevisiae DNA ligase IV can both be found in a complex with a small acidic protein termed XRCC4 in humans and LIF1 in S. cerevisiae ( Critchlow et al., 1997 ; Grawunder et al., 1997 ; Herrmann et al., 1998 ) . A mutation in XRCC4 in mammals or LIF1 in homologous recombination-defective S. cerevisiae leads to loss of NHEJ and hypersensitivity to ionizing radiation. Thus XRCC4 and LIF1 are required for NHEJ, although the exact role of these proteins in DSB repair remains to be established ( Herrmann et al., 1998 ; Jeggo et al., 1995 ). NHEJ also functions in genomic rearrangements, including V(D)J recombination in the generation of immune diversity in mammals, and can function in mating-type switching in S. cerevisiae ( Dynan and Yoo, 1998; Jeggo et al., 1995 ).

The pathway of NHEJ is conserved in eukaryotes as distantly related as S. cerevisiae and humans, and a similar pathway is likely to be found in plants. Here we report the cloning and functional characterization of the first components of the NHEJ pathway from plants. The A. thaliana homologues of DNA ligase IV and XRCC4 proteins are shown to interact, and AtLIG4 is demonstrated to encode a functional DNA ligase. Transcription of the A. thaliana DNA ligase IV gene is found to be inducible by γ-irradiation, consistent with a role in DSB repair.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Cloning and sequence analysis of Arabidopsis DNA ligase IV

Analysis of the public sequence databases for plant homologues of human and S. cerevisiae DNA ligase IV identified a 127-bp region of a 449-bp A. thaliana genomic sequence ( GenBank accession number B22575) which encoded a conserved motif CV/ILDGEM found in DNA ligase IV from S. cerevisiae and humans ( Teo and Jackson, 1997). Primers were designed to this 127 bp region, and a full-length cDNA clone (AtLIG4, 3904 bp, GenBank accession number AF233527) isolated by RACE-PCR. The full-length genomic sequence ( GenBank accession number AB023042, bases 67795–61017) was contained in an A. thaliana chromosome 5 clone subsequently submitted to the database, and the 25 exons within this clone are shown aligned to the cDNA isolated via RACE-PCR ( Figure 1a). The AtLIG4 cDNA contained an open reading frame of 3657 bp, predicted to encode a protein of 1219 amino-acid residues (theoretical molecular mass 138 kDa). The AtLIG4 protein showed 29% identity (48% similarity) with human DNA ligase IV, and 23% identity (43% similarity) with S. cerevisiae DNA ligase IV. Bands at the size predicted by analysis of the genomic sequence were obtained by Southern analysis of A. thaliana DNA digests ( Figure 1b), suggesting that DNA ligase IV is a single-copy gene in the A. thaliana genome.

image

Figure 1. Genomic organization and cDNA cloning strategy of A. thaliana DNA ligase IV.

(a) Primers were designed to a 127 bp region of A. thaliana genomic DNA (accession number B22575) as indicated by the arrows. 5′ and 3′ cDNA ends were amplified by RACE-PCR. The full-length genomic sequence was published during sequencing of chromosome 5 (accession number AB023042, bases 67795–61017) and is shown schematically with exons as boxes, and 3′ and 5′ untranslated regions as shaded black boxes.

(b) Southern analysis of A. thaliana DNA ligase IV. An 879 bp probe corresponding to nucleotides 1937–2816 of AtLIG4 was used in Southern analysis of A. thaliana genomic DNA. Genomic DNA was incubated overnight at 37°C with BamHI (lane 1), HindIII (lane 2), SalI (lane 3) or XbaI (lane 4), separated on a 0.8% agarose gel, transferred to nylon membrane and analysed by hybridization.

Download figure to PowerPoint

The predicted amino-acid sequence of AtLIG4 contained a DNA ligase catalytic domain and tandem BRCA C-terminal (BRCT) domains, similar to the structure of human and S. cerevisiae DNA ligase IV proteins ( Figure 2). BRCT domains are involved in protein–protein interactions, and are present in a number of proteins involved in DNA repair and cell cycle control ( Callebaut and Mornon, 1997) . Human and S. cerevisiae DNA ligase IV BRCT domains are the sites of XRCC4 and LIF1 binding, respectively ( Critchlow et al., 1997 ; Herrmann et al., 1998 ). A. thaliana DNA ligase IV also has an additional acidic, serine/threonine-rich 310 amino-acid (34 kDa) C-terminal domain not found in the human and S. cerevisiae DNA ligase IV proteins ( Figure 2). This region is encoded by a single exon and has a high number of charged residues (37%) and a pI of 5.5 ( Figures 1a and 2). Database searches for sequences similar to this domain identified a range of highly charged and/or serine/threonine-rich proteins, including cytoskeletal proteins and nucleic acid-binding proteins which contained domains resembling this region (data not shown). The functional significance of this unique DNA ligase IV domain remains unclear, although the nuclear localization signals and similarity to nucleic acid-binding proteins suggests a role in subcellular localization and/or DNA binding.

image

Figure 2. Alignment of DNA ligase IV from A. thaliana, humans and S. cerevisiae.

Protein sequences of A. thaliana DNA ligase IV (GenBank accession number AF233527), human DNA ligase IV (GenBank accession number X83441) and S. cerevisiae DNA ligase IV (GenBank accession number CAA99193) were aligned using clustal w ( Thompson et al., 1994 ) and the alignment annotated to give black boxes around identical residues and grey boxes around conservative substitutions aligned in two or more sequences. BRCT domains were identified by comparison with the Prosite database ( http://www.isrec.isb-sib.ch/) and are indicated by black bars. The putative nuclear localization signals in A. thaliana DNA ligase IV are shown in white boxes, and the active site lysine residue is indicated by an asterisk.

Download figure to PowerPoint

AtLIG4 encodes an ATP-dependent DNA ligase

The A. thaliana DNA ligase IV catalytic domain (AtLIG4-CD, nucleotides 59–1862) was overexpressed in Escherichia coli with an N-terminal 6×His tag, and the protein purified to homogeneity by immobilized metal-affinity chromatography. Purified AtLIG4-CD was assayed for DNA ligase activity using three different substrates ( Figure 3a,b). Substrates in which oligo dT was hybridized to either poly dA or poly rA were successfully ligated by AtLIG4-CD, forming mostly dimers and trimers of the oligo dT ( Figure 3a). Although the ligation activity of purified AtLIG4-CD initially appeared to be independent of ATP, pretreatment of these ligase preparations with pyrophosphate (5 m m) abolished joining activity which could be restored only on addition of ATP, thus demonstrating ATP dependence ( Figure 3a).

image

Figure 3. AtLIG4 encodes a functional DNA ligase.

(a) Purified AtLIG4-CD was assayed for ATP-dependent DNA ligase activity using radiolabelled oligo dT (OdT) hybridized to either poly dA (PdA) or poly A (PrA) as substrates. AtLIG4-CD ligase activity was determined either with (+) or without (–) pretreatment of AtLIG4-CD with pyrophosphate (PPi, 5 m m) in the presence (+) or absence (–) of ATP (1 m m), as indicated.

(b) ATP-dependent ligase activity of AtLIG4-CD either with (+) or without (–) pretreatment with pyrophosphate (PPi, 5 m m) was assayed using radiolabelled oligo A (OrA) hybridized to poly dT (PdT) in the presence (+) or absence (–) of ATP (1 m m).

Download figure to PowerPoint

When oligo rA hybridized to poly dT was used in assays as a substrate for DNA ligase IV, then the oligo rA was ligated to a lesser extent than oligo dT/poly dA, with much of the substrate remaining unligated ( Figure 3b). That this joining activity is ATP-dependent is indicated by a small amount of r(A)2n ligated product in assays using pyrophosphate-treated DNA ligase IV in the presence of ATP but the absence of such a product in assays without ATP ( Figure 3b). Thus the substrate specificity of A. thaliana DNA ligase IV differs from that of human and S. cerevisiae DNA ligase IV, as these ligase IV enzymes have no activity with oligo rA/poly dT substrates, and only human DNA ligase III is able to join this substrate effectively ( Robins and Lindahl, 1996; Tomkinson and Mackey, 1998).

Identification of an Arabidopsis homologue of mammalian XRCC4

DNA ligase IV from human and S. cerevisiae cells co-purifies as a complex with XRCC4 and LIF1, respectively ( Critchlow et al., 1997 ; Herrmann et al., 1998 ). Both XRCC4 and LIF1 are small acidic proteins, but they share only limited amino-acid sequence similarity. Database analysis designed to identify an A. thaliana homologue of these proteins revealed a genomic clone located on chromosome 1 which displayed significant sequence similarity to human XRCC4. The ORF within this clone had been determined (T1F9.10, GenBank accession number AAC13900) and was predicted to encode a protein of unknown function. Our attempts to clone the corresponding cDNA by RT-PCR using RNA from flowers, siliques, and vegetative tissue from seedlings and mature plants were unsuccessful, suggesting either that T1F9.10 is not expressed, or that expression is highly regulated in A. thaliana tissues. The genomic sequence of a clone from A. thaliana chromosome 3 ( GenBank accession number AB026655) contained a sequence similar to T1F9.10 (91% identity at the nucleotide level in a gapped alignment). The presence of putative introns in this clone reduced the superficial similarity to XRCC4, explaining why this sequence was not identified in the initial database search for A. thaliana XRCC4 homologues. A cDNA containing the ORF of 792 bp (designated AtXRCC4, GenBank accession number AF233528) was obtained by RACE-PCR using RNA isolated from aerial tissue of 3-week-old A. thaliana. The ORF of this cDNA clone encoded a protein of 264 amino-acid residues (predicted molecular mass 30 kDa, pI 5.7) designated AtXRCC4, and represented a putative A. thaliana XRCC4 homologue. AtXRCC4 showed the highest sequence similarity to human XRCC4 (25% identity, 43% similarity) but did not show any overall significant similarity to LIF1. However a closer inspection of the amino-acid sequences of AtXRCC4, human XRCC4 and LIF1 ( Figure 4) reveals sequence similarity between AtXRCC4 and the region of LIF1 (residues 190–260) involved in the interaction with the BRCT domain of S. cerevisiae DNA ligase IV (21% identity, 48% similarity over 73 amino acids; Herrmann et al., 1998 ).

image

Figure 4. Alignment of A. thaliana XRCC4, human XRCC4 and S. cerevisiae LIF1.

Protein sequences were aligned using clustal w ( Thompson et al., 1994 ) and the alignment annotated to give black boxes around identical residues and grey boxes around conservative substitutions aligned in two or more sequences. Residues 190–260 of LIF1 were identified by Herrmann et al. (1998) as the region responsible for interaction with DNA ligase IV.

Download figure to PowerPoint

AtXRCC4 interacts with DNA ligase IV

To determine whether AtXRCC4 had a homologous function to LIF1 and human XRCC4, the possibility of molecular interaction between A. thaliana DNA ligase IV and AtXRCC4 was investigated using two approaches. Using the yeast two-hybrid system, A. thaliana DNA ligase IV BRCT (LIG4-BRCT) domains were expressed as a fusion protein with the GAL4 DNA binding domain (GAL4DB; Figure 5a), and A. thaliana XRCC4 was expressed as a fusion protein with the GAL4 transcriptional activation domain (GAL4AD). As controls, A. thaliana DNA ligase IV catalytic domain (LIG4-CD) was expressed as a GAL4DB fusion ( Figure 5a), and the interacting proteins SNF1–GAL4AD and SNF4–GAL4DB were included. Interaction between proteins was measured by induction of β-galactosidase expression in S. cerevisiae. Interaction between LIG4-BRCT domains and XRCC4 resulted in induction of β-galactosidase expression similar to that of the positive control (SNF1 and SNF4; Fields and Song, 1989). The negative controls showed no interaction between LIG4-CD and XRCC4, SNF4 and XRCC4, or SNF1 and LIG4-BRCT domains. Thus, yeast two-hybrid analysis demonstrated strong binding of AtXRCC4 to the tandem BRCT domains of A. thaliana DNA ligase IV, but no interaction with the catalytic domain of this DNA ligase IV ( Figure 5b).

image

Figure 5. A thaliana DNA ligase IV interacts with A. thaliana XRCC4.

(a) DNA ligase IV protein structure and constructs used in this study. Schematic alignment of A. thaliana DNA ligase IV (AtLIG4) with human (Homo sapiens) DNA ligase IV (HsLIG4) and yeast (Saccharomyces cerevisiae) DNA ligase IV (ScLIG4). Proteins are shown aligned at the DNA ligase active site as indicated (amino-acid consensus sequence KXDGXR, Tomkinson and Levin, 1997). BRCT domains are depicted as cross-hatched boxes, and the positions of nuclear localization signals (NLS) are shown. The N-terminal catalytic domain (AtLIG4-CD) and the BRCT domain (AtLIG4-BRCT) constructs are shown.

(b) Yeast two-hybrid analysis of DNA ligase IV interaction with XRCC4. A. thaliana DNA ligase IV catalytic domain (LIG4-CD), DNA ligase IV BRCT (LIG4-BRCT) domains and XRCC4 were expressed as fusion proteins with GAL4 DNA-binding domain (GAL4DB) or GAL4 transcriptional activation domain (GAL4AD) as indicated. Control hybrid proteins SNF1-GAL4AD and SNF4-GAL4DB were included. Interaction between proteins was measured by induction of β-galactosidase expression in S. cerevisiae strain Y190.

(c) Interaction between DNA ligase IV and XRCC4 on affinity column chromatography. A. thaliana DNA ligase IV BRCT domains were overexpressed as a poly histidine-tagged protein in E. coli, and applied to a nickel affinity column. The column was washed and the soluble fraction of an E. coli lysate containing overexpressed calmodulin-binding protein (CBP)-tagged XRCC4 was applied to the column. The column was washed extensively, and a step gradient of 200–1000 m m NaCl passed down the column before a final elution of the histidine-tagged BRCT domains with EDTA (25 m m). Fractions were analysed by Western analysis using biotinylated calmodulin and alkaline phosphatase-conjugated streptavidin for the detection of CBP-tagged XRCC4.

Download figure to PowerPoint

These results were confirmed by affinity-column chromatography. The full-length AtXRCC4 cDNA was overexpressed as a calmodulin-binding protein fusion protein in E. coli. Western analysis showed that recombinant calmodulin-binding protein tagged AtXRCC4 (AtXRCC4-CBP) migrated as a 42 kDa protein which is larger than the predicted size of 34 kDa (including the affinity tag). Slow migration in SDS-PAGE was also observed for LIF1 and human XRCC4, and may be due to the hydrophilic nature of the protein reducing SDS binding ( Critchlow et al., 1997 ; Herrmann et al., 1998 ). Interaction between recombinant AtXRCC4 and 6×His-tagged DNA ligase IV BRCT domains bound to a nickel-affinity column was investigated. AtXRCC4-CBP was passed down the BRCT domain-bound affinity column and eluted with a step NaCl gradient, with a final elution with EDTA to release the 6×His protein. XRCC4 binding to DNA ligase IV BRCT domains was stable to at least 1 m NaCl, and both proteins were only co-eluted from the column in the EDTA wash ( Figure 5c).

Expression patterns of DNA ligase IV in Arabidopsis and induction by γ-irradiation

DNA ligase IV was expressed in all the A. thaliana tissues studied, with the highest levels of transcript expression in young flowers at the time of bud opening and in roots ( Figure 6a). Ubiquitous expression of DNA ligase IV is consistent with a role in DNA repair, although ligase IV may also function in mitosis and/or meiosis.

image

Figure 6. Tissue specificity of DNA ligase IV expression and response of DNA ligases I and IV and XRCC4 expression to γ-irradiation.

(a) Tissue specificity of DNA ligase IV expression. Northern analysis of total RNA using an 879 bp probe corresponding to nucleotides 1937–2816 of AtLIG4 was quantified with a BAS2000 Phosphorimager, and results were normalized to the hybridization signal of a 28S rRNA probe.

(b) Expression of DNA ligase I, IV and XRCC4 in response to γ-irradiation dose. A. thaliana plants were exposed to a Co60 source for 20 min at dose rates from 0.015–0.5 Gy min−1. Total RNA was isolated from aerial parts of 3-week-old A. thaliana plants 3 h after γ-irradiation at the dose indicated. Transcript levels of ●, DNA ligase IV; ▴, XRCC4; ▪, DNA ligase I were determined by Northern analysis and quantified by normalized densitometry.

(c) Induction of DNA ligase I, IV and XRCC4 expression with time after γ-irradiation of 1 Gy under the same conditions as described for (b). Total RNA was isolated from aerial parts of 3-week-old A. thaliana plants at the time indicated. Transcript levels of ●, DNA ligase IV; ▴, XRCC4; ▪, DNA ligase I were determined by Northern analysis and quantified by normalized densitometry.

Download figure to PowerPoint

Northern analysis and densitometry were also used to quantify expression levels of DNA ligase IV, XRCC4 and DNA ligase I in aerial tissues of 3-week-old A. thaliana plants in response to γ-irradiation, which causes a range of DNA damage in cells including DSBs ( Ward, 1988). DNA ligase IV expression increased in response to γ-radiation dose with a 3.6-fold increase in transcript levels in aerial tissues 3 h after a 10 Gy dose ( Figure 6b). In contrast, DNA ligase I and XRCC4 transcript levels showed only slight increases (60%) with the same treatment. There was an initial lag period of 2–3 h after induction of DSBs by γ-irradiation before DNA ligase IV transcript levels started to increase, shown in Figure 6(c). DNA ligase IV transcript levels reached 5.6 times basal levels 6 h after irradiation treatment. Both XRCC4 and DNA ligase I transcript levels showed a threefold increase 6 h after γ-irradiation, but the induction of transcription of both XRCC4 and DNA ligase I was delayed relative to that of DNA ligase IV. The 2–3 h lag in induction of DNA ligase IV transcription probably represents the time taken for DNA damage detection and induction of intracellular signalling pathways, resulting in increased transcription of DNA ligase IV. The greater delay of XRCC4 and DNA ligase I transcript induction may originate from a delay in any of these processes. The increase in XRCC4 and DNA ligase I transcript levels may represent a general up-regulation of DNA repair gene expression following extensive DNA damage, and contrasts with the large and rapid increase in DNA ligase IV transcript levels on γ-irradiation. In addition, XRCC4 protein levels may not be limiting in the NHEJ pathway, and so induction of XRCC4 expression on γ-irradiation may not be required for the increase in DSB repair following DNA damage.

Regulation of expression of XRCC4 and DNA ligases I and IV by UVB and white light

Ultraviolet B (UVB) irradiation is the most DNA-damaging component of sunlight reaching the Earth's surface, and causes the formation of pyrimidine dimers and oxidative DNA damage including single-strand breaks in the plant genome. In plants, pyrimidine dimers are removed from damaged DNA by light-dependent repair (photoreactivation) of cyclobutane pyrimidine dimers and 6-4 photoproducts (6-4PPs) by photolyase enzymes which directly reverse the DNA damage in an error-free manner. In the absence of visible light, independent base and nucleotide excision repair pathways can be used by plants (so-called ‘dark repair’; Taylor et al., 1996 ). UVB irradiation of 3-week-old A. thaliana plants in the dark has little effect on transcript levels of XRCC4 and DNA ligase IV in aerial tissue, whereas DNA ligase I transcript levels show a twofold increase after 9 h of UVB irradiation ( Figure 7a). Irradiation of A. thaliana plants by white light supplemented with UVB for 8 h results in increases in DNA ligase I and IV transcript levels of 3.5-fold and 2.2-fold, respectively, but no significant increase in XRCC4 transcript levels ( Figure 7b). Comparison of transcript levels in UVB-treated plants in the presence and absence of white light treatments suggests that the increase in transcript levels observed in the presence of white light supplemented with UVB is largely attributable to the white light component of the irradiation, rather than the UVB irradiation. White light induction of DNA ligase I transcription in A. thaliana plants has been reported previously ( Taylor et al., 1998a ), and a similar role for white light in the induction of DNA ligase IV transcription is indicated in the present study. Thus UVB irradiation, which causes single-strand DNA breaks both directly and through the excision repair pathways, induces DNA ligase I but not ligase IV expression ( Figure 7a). This is in contrast to γ-irradiation, which causes both double- and single-strand DNA breaks and leads to an irradiation-dose-dependent induction of DNA ligase IV ( Figure 6b). Induction of DNA ligase IV by γ-irradiation but not by UVB irradiation is consistent with a role of this enzyme in DSB repair.

image

Figure 7. Expression of DNA ligases I and IV and XRCC4 in response to UVB irradiation and white light.

(a) Expression of DNA ligase I, IV and XRCC4 in response to UVB irradiation in the dark. Total RNA was isolated from aerial parts of 3-week-old A. thaliana plants at the time indicated after the onset of continuous UVB irradiation (1.25 W m−2) after 8 h in the dark. Transcript levels of ●, DNA ligase IV; ▴, XRCC4; ▪, DNA ligase I were determined by Northern analysis and quantified by normalized densitometry.

(b) Expression of DNA ligase I, IV and XRCC4 in response to UVB irradiation in the light. UVB irradiation (1.25 W m−2) commenced at the onset of the light cycle, and total RNA was isolated from aerial parts of 3-week-old A. thaliana plants at the time of irradiation as indicated. Transcript levels of ●, DNA ligase IV; ▴, XRCC4; ▪, DNA ligase I were determined by Northern analysis and quantified by normalized densitometry.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

NHEJ, rather than homologous recombination, is the major pathway for DSB repair in plants ( Gorbunova and Levy, 1999). In S. cerevisiae and humans, components of the NHEJ pathways show a significant degree of sequence similarity despite the large phylogenetic distance between the species. Here we have identified the first plant homologues of the S. cerevisiae and human NHEJ proteins DNA ligase IV and XRCC4/LIF1, and demonstrated their interaction.

In mammals three separate DNA ligase genes have been identified encoding DNA ligases I, III and IV, which have distinct roles in cellular metabolism. Homologues of DNA ligases I and IV are found in S. cerevisiae, and analysis of the complete S. cerevisiae genomic sequence indicates that S. cerevisiae has only these two forms of DNA ligase, and no homologue of DNA ligase III ( Teo and Jackson, 1997). However, the S. cerevisiae DNA ligase I transcript contains alternative translation start sites, encoding nuclear and mitochondrial forms of the enzyme ( Willer et al., 1999 ). Sequence analysis of human DNA ligase III suggests that the gene may have arisen early in vertebrate evolution ( Teo and Jackson, 1997) and no plant homologue of DNA ligase III has yet been identified, although a novel putative DNA ligase has been identified in Oryza sativa and A. thaliana (C.E. West, unpublished results).

Database analysis has identified a putative A. thaliana DNA ligase IV gene. Functional analysis of the recombinant protein has demonstrated that this A. thaliana DNA ligase IV catalyses ATP-dependent joining of (dT)16 oligomers hybridized to either poly dA or poly rA, and weak activity on (rA)16 oligomers hybridized to poly dT. The protein encoded by the corresponding cDNA is structurally similar to the S. cerevisiae and human DNA ligase IV proteins. The A. thaliana DNA ligase IV protein has an N-terminal DNA ligase catalytic domain which contrasts with the position of this domain in DNA ligase I of mammals, S. cerevisiae and A. thaliana, where it is located at the C-terminus ( Tomkinson and Mackey, 1998). The A. thaliana DNA ligase IV contains tandem BRCT domains, also found in S. cerevisiae and human DNA ligase IV, but in addition has a C-terminal domain which is unique to A. thaliana DNA ligase IV. This C-terminal domain is similar to serine-rich and charged regions of cytoskeletal proteins and nucleic acid-binding proteins, the latter suggesting a possible role of this domain in DNA binding. Alternatively, as the regions of sequence similarity contain numerous putative nuclear localization sequences, this domain may function in subcellular localization of DNA ligase IV. The BRCT domains that are present in a range of proteins which function in DNA repair and cell cycle control are involved in protein–protein interactions. In human and S. cerevisiae DNA ligase IV, the BCRT domains present in these proteins are responsible for the strong interaction of DNA ligase IV with XRCC4 and LIF1, respectively.

We have identified a putative A. thaliana XRCC4 homologue and demonstrated that this protein possesses properties similar to human XRCC4. Specifically, we have shown that the A. thaliana XRCC4 protein interacts strongly with the A. thaliana DNA ligase IV BRCT domains, but not with the DNA ligase IV N-terminal catalytic domain. The strength of this A. thaliana XRCC4-DNA ligase IV interaction was shown by its resistance to dissociation by high salt concentrations. Protein–protein complexes that are stable in high salt concentration may be indicative of hydrophobic interactions between interacting proteins. These observations suggest that the A. thaliana DNA ligase IV/XRCC4 complex appears to be homologous to the human DNA ligase IV/XRCC4 complex and the S. cerevisiae DNA ligase IV/LIF1 complex.

Mammalian XRCC4 was first identified by complementation of X-ray-sensitive hamster cell lines, and human DNA ligase IV co-purifies as a complex with XRCC4. The function of XRCC4 is unclear, although it is known to have an essential role in NHEJ. XRCC4 can stimulate human DNA ligase IV activity in vitro fivefold, and this stimulation may be due to an increase in the rate of the adenylation step of catalysis rather than localization of DNA ligase IV to the DNA substrate by XRCC4 ( Grawunder et al., 1997 ; Modesti et al., 1999 ) . In vivo, both XRCC4 and LIF1 stabilize DNA ligase IV protein levels whereas, in contrast, in XRCC4 and LIF1 mutants DNA ligase IV protein levels are undetectable by Western analysis ( Bryans et al., 1999 ; Herrmann et al., 1998 ). This may indicate a role of XRCC4 in the regulation of DNA ligase protein turnover, possibly by XRCC4 conferring resistance to proteolytic digestion on DNA ligase IV via complex formation. Human XRCC4 binds DNA co-operatively in vitro independent of DNA sequence, but with higher affinity for nicked or linear DNA than closed circular DNA ( Modesti et al., 1998 ). The N-terminal (1–28 amino acids) and central regions (168–200) of human XRCC4 are required for DNA binding, whilst the central region (168–200) is also required for self-interaction, suggesting that self-association of XRCC4 is necessary for DNA binding ( Modesti et al., 1998 ). XRCC4 may bind to other proteins of the NHEJ complex in vivo, including interaction with Ku and DNA-PKcs in mammals ( Leber et al., 1998 ), and evidence suggests that XRCC4 serves to localize DNA ligase IV to the site of DSB repair. The localization of XRCC4 to the site of DSB repair may be subject to regulation as phosphorylation of XRCC4 abolishes its ability to bind DNA, although interaction with DNA ligase IV is unaffected ( Modesti et al., 1998 ).

DNA ligase IV transcript levels in A. thaliana showed a rapid increase within a few hours of γ-irradiation of tissues, and transcript levels 3 h after irradiation were γ-irradiation dose-dependent. Gamma radiation produces a range of DNA lesions, including single- and double-strand breaks estimated to occur at a ratio of 20 : 1 in nuclear DNA ( Plumb et al., 1999 ). DNA ligase I and XRCC4 showed a delayed increase in transcript levels on γ-irradiation and were not induced to the same extent as DNA ligase IV. A different pattern of expression was found upon UVB irradiation of A. thaliana in the dark, where photoreactivation will be inactive and any DNA damage incurred will, of necessity, be repaired by the light-independent excision repair pathways. DNA ligase I transcript levels increased with time of UVB-irradiation, which is in contrast to XRCC4 and DNA ligase IV expression which exhibited little change. These observations are consistent with the involvement of DNA ligase IV in DSB repair (specific induction by γ-irradiation) and with DNA ligase I functioning in excision repair pathways (induction by UVB in the dark) in A. thaliana. However, expression of DNA ligases I and IV was induced when UVB irradiation was supplemented with white light. The reason for this induction by white light is unclear. White light regulation of DNA ligase I expression has been suggested previously, and expression of DNA ligase I may also be linked to DNA replication ( Taylor et al., 1998a ).

The A. thaliana DNA ligase IV and XRCC4 proteins are the first plant homologues of the S. cerevisiae and human NHEJ proteins to be characterized, whilst potential A. thaliana homologues of other components of the NHEJ process, for example MRE11 and RAD50, can be found in the public databases. The best characterized repair pathway in plants to date is photoreactivation of UV-induced DNA damage, and photolyases specific for both cyclobutane pyrimidine dimers and 6-4 photoproducts have been cloned ( Britt, 1999; Vonarx et al., 1998 ). The study of other plant DNA-repair pathways lags some way behind photoreactivation, although representatives of both base and nucleotide excision-repair pathways and a mismatch-repair pathway have been characterized in plants ( Britt, 1999; Vonarx et al., 1998 ). Our knowledge of the enzymology of DNA repair pathways and mechanisms is limited in part by the difficulty of applying molecular genetic techniques to plants, although completion of the A. thaliana genomic sequencing programme in combination with T-DNA and transposon tagging should now combine to allow rapid advances.

The presence of DNA ligase IV and XRCC4 homologues in higher plants, along with identification of sequences representing probable components of the NHEJ process in the A. thaliana genome sequencing programme, is indicative of a high degree of conservation of the fundamental mechanisms of NHEJ amongst all eukaryotes.

The identification and molecular characterization of key components of NHEJ in plants should now begin to provide us with the detailed knowledge necessary to understand the maintenance of genome stability and gene flow in plants.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Plant material and growth conditions

A. thaliana (genotype Landsberg erecta) seeds were sown directly onto damp Levington's M2 compost. Light-grown plants were raised in an environmental growth chamber (SGC970/C/HQI, Sanyo-Gallenkamp, Loughborough, UK) under controlled conditions of constant humidity (70%), with 16 h light and 8 h dark cycles at 20°C for the period stated. Visible light was provided by a combination of high-intensity discharge lamps (Osram, Light Source Supplies, Bishop's Stortford, UK) and tungsten lamps. The quantum flux density was measured daily 2 h into the photoperiod, and was equivalent to 100 μmol m−2 s−1 photosynthetically active radiation (PAR) for light-grown tissue. Dark-grown tissue was raised under identical conditions except for the presence of light. UVB irradiation was performed either in the absence of other light sources, or in the presence of white light. Two UV-emitting tubes (Philips TL40/12, Philips, The Netherlands) were present in the growth chambers and were wrapped in cellulose acetate (0.13 mm thick) to filter shorter UVC and UVB wavelengths below 295 nm which are not present in natural sunlight. The cellulose acetate was changed at the start of each experiment.

Molecular cloning

DNA procedures and bacterial manipulations were performed using established protocols ( Sambrook et al., 1989 ) unless otherwise stated. Plasmid DNA was prepared on an analytical scale using Qiagen columns according to the manufacturer's instructions (Qiagen , Crawley, UK).

Reverse transcription was performed using AMV reverse transcriptase (Roche, Sussex, UK), RT-PCR using either Taq DNA polymerase or the Expand PCR kit (Roche), and RACE-PCR was performed using the 3′/5′ RACE-PCR kit (Roche) according to the manufacturer's instructions. PCR products were cloned using a TOPO-TA cloning kit (Invitrogen , Gröningen, Netherlands) and plasmid DNA prepared using spin columns (Qiagen) prior to fluorescent dye-terminator DNA sequencing (Perkin Elmer, Warrington, UK) and analysed on an Applied Biosystems (Warrington, UK) automated sequencer. Oligonucleotide primers were purchased from Amersham. The E. coli strain TOPO10F′ (Invitrogen, genotype F′ {lacIqTn10(TetR)} mcrA Δ(mrr-hsdRMS-mcrBC)Φ80lacZΔM15 ΔlacX74recA1araD139 Δ(ara-leu)7697 galU galK rpsL (StrR) endA1 nupG) was used for routine cloning.

Overexpression of AtLIG4-CD, AtLIG4-BRCT and XRCC4-CBP

AtLIG4-CD (nucleotides 59–1862) was cloned into the plasmid pET30b (Novagen, Nottingham, UK) generating an N-terminal 6×His fusion protein. The AtLIG4-BRCT domain (nucleotides 1937–2816) was cloned into pET30b generating both N- and C-terminal 6×His tags. AtXRCC4 was cloned into the plasmid pCal-c (Stratagene, Amsterdam, Netherlands) and was expressed with a C-terminal calmodulin-binding protein (CBP) tag. Expression of AtLIG4-BRCT and AtXRCC4 was induced for 3 h in E. coli strain BL21(DE3)pLysS (Promega, Southampton, UK) by the addition of isopropylthiogalactoside (1 m m). Bacteria were recovered by centrifugation, resuspended in RS buffer (50 m m Tris–Cl pH 7.5, 50 m m NaCl, 10 m m imidazole, 5% (v/v) glycerol, 0.1% (v/v) Triton X100) and lysed by freeze–thawing and sonication. The extract was cleared by centrifugation at 13 000 g for 10 min. 6 × His-tagged AtLIG4-CD was applied to a 5 ml nickel-chelating Sepharose fast-flow affinity chromatography column (Pharmacia, Little Chalfont, UK) , washed with RS buffer containing 100 m m imidazole, and eluted in RS buffer containing 500 m m imidazole. His-tagged protein was detected using anti 6×His antibody (Sigma), followed by alkaline phosphatase coupled secondary antibody and colour development according to the manufacturer's instructions.

DNA ligase assay

DNA ligase activity of AtLIG4-CD was assayed by ligation of oligonucleotides as described previously ( Taylor et al., 1998b ). Pyrophosphate treatment of purified A. thaliana DNA ligase IV catalytic domain was performed as described previously ( Riballo et al., 1999 ). Pyrophosphate was removed by buffer exchange on a 2.5 ml void volume Sephadex G-25 PD10 column (Pharmacia) prior to assaying for DNA ligase activity.

Isolation of total RNA and Northern analysis

RNA was isolated using either the SV total RNA isolation kit (Promega) or a modified version of the method of Knight and Gray (1994). Fresh tissue (500 mg) was ground in liquid nitrogen, added to 4 ml extraction buffer (100 m m Tris–HCl pH 8.5, 100 m m NaCl, 20 m m EDTA, 1% (w/v) sodium lauryl sarcosinate, 1.7% (w/v) diethyldithiocarbamate), and vortexed. The solution was extracted once with water-saturated phenol (Biogene, Bedfordshire, UK), and twice with chloroform. Nucleic acids were then precipitated and resuspended in RNase-free water. LiCl was added to a final concentration of 2 m, and RNA precipitated for 4 h on ice. RNA was recovered by centrifugation (10 000 g for 10 min) and washed with ice-cold RNase-free water.

RNA was separated by formaldehyde agarose gel electrophoresis and blotted onto nylon membrane (Hybond-N, Amersham), prehybridized and hybridized at 42°C using standard methods ( Sambrook et al., 1989 ). DNA fragments were labelled by random hexanucleotide priming with [α-32P]dCTP (ICN Cosa Mesa, USA) and the Klenow fragment of E. coli DNA polymerase (Stratagene), and separated from unincorporated radiolabel using Sephadex G-50 columns (Nick columns, Pharmacia). Blots were washed twice at room temperature in 2 × SSPE ( Sambrook et al., 1989 ), and once at 42–65°C in 0.1 × SSPE before autoradiography or analysis using a PhosphoImager (Fuji, Tokyo).

Isolation of genomic DNA and Southern analysis

A. thaliana tissue was harvested and ground to a fine powder in liquid N2. DNA was isolated using the method of Doyle and Doyle (1990) and Southern analysis performed as described previously ( Taylor et al., 1998a ).

Yeast two-hybrid system

Two-hybrid analysis was performed as described by Fields and Song (1989) and Durfee et al. (1993) . Nucleotides 1937–2816 encoding the AtLIG4-BRCT domain were cloned into the plasmid pAS.1 ( Durfee et al., 1993 ) to create a GAL4–DNA binding domain fusion protein. AtXRCC4 was cloned into the plasmid pGAD-C1 ( James et al., 1996 ) to create a GAL4–activation domain fusion. S. cerevisiae strain Y190 (genotype MATa, ura3-52, his3-200, ade2-101, lys2-801, trp1-901, leu2-3, 112, gal4Δ, gal80Δ, cyhr2, LYS2::GAL1UAS-HIS3TATA-HIS3, URA3::GAL1UAS-GAL1TATA-lacZ) was used for interaction studies, and plasmids pSE1111 and pSE1112 expressing the SNF1–GAL4–DNA binding domain and the SNF4–GAL4–activation domain, respectively, were used as controls ( Durfee et al., 1993 ). S. cerevisiae was transformed using a lithium acetate method as described previously ( Soni et al., 1993 ).

Column chromatography

His-tagged AtLIG4-BRCT was applied to nickel-chelating Sepharose fast-flow affinity resin (Pharmacia) and washed with RS buffer containing 100 m m imidazole as described above for AtLIG4-CD. Cleared extract of E. coli strain BL21(DE3)pLysS expressing AtXRCC4-CBP was applied to the column-bound AtLIG4-BRCT and the column washed with RS buffer containing NaCl (0.2–1 m). Finally nickel-bound 6×His AtLIG4-BRCT was eluted by addition of EDTA (25 m m). CBP-tagged protein was detected by Western blotting using biotinylated calmodulin (Stratagene) in the presence of 1 m m CaCl2, followed by alkaline phosphatase-conjugated streptavidin and colour development according to the manufacturer's instructions.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We thank Richard Taylor for the kind gift of substrates for the DNA ligase assay, John Collett for providing the Co60γ-irradiation source, and Lis Mudd for A. thaliana cDNA. This work was supported by a research grant from the European Community (Project Number FAIR5-PL97–3711).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Babiychuk, E., Cottrill, P.B., Storozhenko, S., Fuangthong, M., Chen, Y., O'farrell, M.K., Inzé, D., Kushnin, S. (1998) Higher plants possess two structurally different poly (ADP-ribose) polymerases. Plant J. 15, 635 645.
  • Barnes, D.E., Tomkinson, A.E., Lehmann, A.R., Webster, A.D.B., Lindahl, T. (1992) Mutations in the DNA ligase I gene of an individual with immunodeficiencies and cellular hypersensitivity to DNA damaging agents. Cell, 69, 495 502.
  • Britt, A.B. (1999) Molecular genetics of DNA repair in higher plants. Trends Plant Sci. 4, 20 25.DOI: 10.1016/s1360-1385(98)01355-7
  • Bryans, M., Valenzano, M.C., Stamato, T.D. (1999) Absence of DNA ligase IV protein in XR-1 cells: evidence for stabilisation by XRCC4. Mutation Res. 433, 53 58.
  • Caldecott, K.W., McKeown, C.K., Tucker, J.D., Ljunquist, S., Thompson, L.H. (1994) An interaction between the mammalian DNA repair protein XRCC1 and DNA ligase III. Mol. Cell. Biol. 14, 68 76.
  • Callebaut, I. & Mornon, J.-P. (1997) From BRCA1 to RAP1: a widespread BRCT module closely associated with DNA repair. FEBS, 400, 25 30.
  • Chu, G. (1997) Double strand break repair. J. Biol. Chem. 39, 24097 24100.
  • Critchlow, S.E. & Jackson, S.P. (1998) DNA end joining: from yeast to man. TIBS, 23, 394 402.
  • Critchlow, S.E., Bowater, R.P., Jackson, S.P. (1997) Mammalian DNA double-strand break repair protein XRCC4 interacts with DNA ligase IV. Current Biol. 7, 588 598.
  • Doyle, J.J. & Doyle, J.L. (1990) Isolation of plant DNA from fresh tissue. Focus, 12, 13 15.
  • Durfee, T., Becherer, K., Chen, P.-L., Yeh, S.-H., Yang, Y., Kilburn, A.E., Lee, W.-H., Elledge, S. (1993) The retinoblastoma protein associates with the protein phosphatase type 1 catalytic subunit. Genes Dev. 7, 555 569.
  • Dynan, W.S. & Yoo, S. (1998) Interaction of Ku protein and DNA dependent protein kinase catalytic subunit with nucleic acids. Nucl. Acids Res. 26, 1551 1559.
  • Fields, S. & Song, O. (1989) A novel genetic system to detect protein–protein interactions. Nature, 340, 245 246.
  • Gorbunova, V. & Levy, A.A. (1997) Non-homologous DNA end joining in plant cells is associated with deletions and filler DNA insertions. Nucl. Acids Res. 25, 4650 4657.
  • Gorbunova, V. & Levy, A.A. (1999) How plants make ends meet: DNA double strand break repair. Trends Plant Sci. 4, 263 269.DOI: 10.1016/s1360-1385(99)01430-2
  • Grawunder, U., Wilm, M., Wu, X., Kulesa, P., Wilson, T.E., Mann, M., Lieber, M.R. (1997) Activity of DNA ligase IV stimulated by complex formation with XRCC4 protein in mammalian cells. Nature, 388, 492 495.
  • Herrmann, G., Lindahl, T., Schär, P. (1998) Saccharomyces cerevisiae LIF1: a function involved in DNA double-strand break repair related to mammalian XRCC4. EMBO J. 17, 4188 4198.
  • James, P., Halladay, J., Craig, E.A. (1996) Genomic libraries and a host strain designed for highly efficient two-hybrid selection in yeast. Genetics, 144, 1425 1436.
  • Jeggo, P.A., Taccioli, G.E., Jackson, S.P. (1995) Menage à trois: double strand break repair, V (D) J recombination and DNA-PK. Bioessays, 17, 949 957.
  • Kanaar, R., Hoeijmakers, J.H.J., Van Gent, D.C. (1998) Molecular mechanisms of DNA double-strand break repair. Trends Cell Biol. 8, 483 498.DOI: 10.1016/s0962-8924(98)01383-x
  • Knight, J.S. & Gray, J.C. (1994) Expression of genes encoding the tobacco chloroplast phosphate translocator is not light regulated and is repressed by sucrose. Mol. Gen. Genet. 242, 586 594.
  • Lakshmipathy, U. & Campbell, C. (1999) The human DNA ligase III gene encodes both nuclear and mitochondrial proteins. Mol. Cell Biol. 19, 3869 3876.
  • Leber, R., Wise, T.W., Mizuta, R., Meek, K. (1998) The Xrcc4 gene product is a target for and interacts with the DNA-dependent protein kinase. J. Biol. Chem. 273, 1794 1801.
  • Li, C., Goodchild, J., Baril, E.F. (1994) DNA ligase I is associated with the 21S complex of enzymes for DNA synthesis in HeLa cells. Nucl. Acids Res. 22, 632 638.
  • Liljegren, S.J. & Yanofsky, M.F. (1998) Towards targeted transformation in plant response: targeting Arabidopsis. Trends Plant Sci. 3, 79 80.
  • Mackey, Z.B., Ramos, W., Levin, D.S., Walter, C.A., McCarrey, J.R., Tomkinson, A.E. (1996) An alternative splicing event, which occurs in mouse pachytene spermatocytes, generates a form of DNA ligase III with distinct biochemical properties that may function in meiotic recombination. Mol. Cell. Biol. 17, 989 998.
  • Mengiste, T. & Paszkowski, J. (1999) Prospects for the precise engineering of plant genomes by homologous recombination. Biol. Chem. 380, 749 758.
  • Modesti, M., Hesse, J.E., Gelbert, M. (1999) DNA binding of XRCC4 protein is associated with V (D) J recombination but not stimulation of DNA ligase IV activity. EMBO J. 18, 2008 2018.
  • Plumb, M.A., Smith, G.C.M., Cunniffe, S.M.T., Jackson, S.P. (1999) DNA-PK activation by ionising radiation-induced DNA single strand breaks. Int. J. Radiat. Biol. 75, 553 561.DOI: 10.1080/095530099140195
  • Puchta, H. (1998) Towards targeted transformation in plants. Trends Plant Sci. 3, 77 78.
  • Riballo, E., Critchlow, S.E., Heo, S.-H. et al. (1999) Identification of a defect in DNA ligase IV in a radiosensitive leukaemia patient. Current Biol. 9, 699 702.DOI: 10.1016/s0960-9822(99)80311-x
  • Robins, P. & Lindahl, T. (1996) DNA ligase IV from Hela cell nuclei. J. Biol. Chem. 271, 24257 24261.
  • Salomon, S. & Puchta, H. (1998) Capture of genomic and T-DNA sequences during double-strand break repair in somatic plant cells. EMBO, 17, 6086 6095.
  • Sambrook, J., Fritsch, E.F., Maniatis, T. (1989) Molecular Cloning, a Laboratory Manual. Cold Spring Harbor:Cold Spring Harbor Press
  • Schär, P., Herrmann, G., Daly, G., Lindahl, T. (1997) A newly identified DNA ligase of Saccharomyces cerevisiae involved in RAD52 independent repair of DNA double strand breaks. Genes Dev. 11, 1912 1924.
  • Soni, R., Carmichael, J.P., Murray, J.A. (1993) Parameters affecting lithium acetate mediated transformation of Saccharomyces cerevisiae and development of a rapid and simplified procedure. Curr. Genet. 24, 445 459.
  • Taylor, R.M., Nikaido, O., Jordan, B.R., Rosamond, J., Bray, C.M., Tobin, A.K. (1996) Ultraviolet-B-induced DNA lesions and their removal in wheat (Triticum aestivum L.) leaves. Plant Cell Environ. 19, 171 181.
  • Taylor, R.M., Hamer, M.J., Rosamond, J., Bray, C.M. (1998a) Molecular cloning and functional analysis of the Arabidopsis thaliana DNA ligase I homologue. Plant J. 14, 75 81.
  • Taylor, R.M., Whitehouse, J., Cappelli, E., Frosina, G., Caldecott, K.W. (1998b) Role of the DNA ligase III zinc finger in polynucleotide binding and ligation. Nucl. Acids Res. 26, 4804 4810.
  • Teo, S.H. & Jackson, S.P. (1997) Identification of Saccharomyces cerevisiae DNA ligase IV: involvement in DNA double-strand break repair. EMBO J. 16, 4788 4795.
  • Thompson, J.D., Higgins, D.G., Gibson, T.J. (1994) clustal w: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice . Nucl. Acids Res. 22, 4673 4680.
  • Tomkinson, A.E. & Levin, D.S. (1997) Mammalian DNA ligases. Bioessays, 19, 893 901.
  • Tomkinson, A.E. & Mackey, Z.B. (1998) Structure and function of mammalian DNA ligases. Mutation Res. 407, 1 9.
  • Vonarx, E.J., Mitchell, H.L., Karhikeyan, R., Chatterjee, I., Kunz, B.A. (1998) DNA repair in higher plants. Mutation Res. 400, 187 200.
  • Ward, J.F. (1988) DNA damage produced by ionising radiation in mammalian cells: identities, mechanisms of formation and reparability. Prog. Nucl. Acids Res. Mol. Biol. 35, 95 125.
  • Willer, M., Rainey, M., Pullen, T., Stirling, C.J. (1999) The yeast CDC9 gene encodes both a nuclear and a mitochondrial form of DNA ligase I. Curr. Biol. 9, 1085 1094.DOI: 10.1016/s0960-9822(99)80477-1
  • Wilson, T.E., Grawunder, U., Lieber, M.R. (1997) Yeast DNA ligase IV mediates non-homologous DNA end joining. Nature, 388, 495 498.
  • Wood, R.D. (1996) DNA repair in eukaryotes. Ann. Rev. Biochem. 65, 135 167.

GenBank accession numbers AF233527 (Arabidopsis DNA ligase IV) and AF233528(ArabidopsisXRCC4).