![]()
|
|
||||||||
J. Biol. Chem., Vol. 278, Issue 43, 41631-41635, October 24, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||



From the Institute of Molecular Medicine and Genetics, Medical College of Georgia, Augusta, Georgia 30912
Received for publication, June 18, 2003 , and in revised form, August 8, 2003.
| ABSTRACT |
|---|
|
|
|---|
200-kDa factor that does not correspond to any of the previously described double-strand break repair proteins. The two pathways converge, sharing a common requirement for the DNA ligase IV-XRCC4 complex to catalyze the final step of phosphodiester bond formation. Whereas the Mre11-Rad50-NBS1-dependent pathway does not require, and may be inhibited by, DNA-dependent protein kinase-mediated phosphorylation, the new pathway depends on this phosphorylation for release from a DNA-dependent protein kinase-mediated reaction checkpoint. The existence of two distinct pathways, which are differentially regulated by the DNA-dependent protein kinase, provides a possible explanation for the selective repair defects seen in DNA-dependent protein kinase-deficient mutants. | INTRODUCTION |
|---|
|
|
|---|
In S. cerevisiae, NHEJ also requires the M·R·X complex, composed of MRE11p, RAD50p, and XRS2p (reviewed in Ref. 3). This multifunctional complex remodels and aligns broken DNA ends. The role of the M·R·X complex in other organisms is uncertain. The equivalent complex is not essential for NHEJ in Schizosaccharomyces pombe (4). In vertebrates, the equivalent complex is essential for cell viability in the absence of DNA damage, so its specific role in NHEJ is difficult to assess (57). Naturally occurring mutant alleles of two proteins in the complex, Mre11 and NBS1 (functional equivalents of yeast MRE11p and XRS2p, respectively), do not cause specific NHEJ defects, although they are associated with human genetic instability syndromes (810).
In vertebrates, NHEJ requires several proteins with no apparent orthologs in lower eukaryotes. These include the DNA-dependent protein kinase catalytic subunit (DNA-PKcs), which interacts with Ku and DNA to form an active protein kinase complex, and Artemis, which processes DNA hairpin ends. Human NHEJ also requires a new, as yet unidentified gene, which apparently does not correspond to any of the genes identified in yeast (10). The requirement for additional proteins in vertebrates may reflect the larger genome size, where the increased likelihood of multiple, simultaneous breaks demands a more efficient NHEJ system. In yeast, NHEJ is a secondary mechanism of repair that comes into play only when homologous recombination is disabled.
Understanding the relationship between vertebrate-specific NHEJ proteins, especially DNA-PKcs, and the more primitive NHEJ system in lower eukaryotes is a central challenge in the field. Do all of the proteins cooperate in a single pathway, or are there multiple pathways involving distinct sets of proteins, perhaps targeted at repair of different types of DSBs? It is notable that phenotypes of mutants in vertebrate NHEJ components vary in severity. Targeted disruption of genes encoding murine L4 and X4 causes death in utero (11, 12). Disruption of genes encoding Ku produces viable mice that exhibit radiation sensitivity, defective V(D)J recombination, progeria, and dwarfism (13, 14). Disruption of the gene encoding DNA-PKcs produces mice that are normal except for radiation sensitivity and selective repair defects (15, 16). The graded severity of mutant phenotypes is consistent with the idea of branched NHEJ pathways with some requirements that are distinct and others that are common.
Here we apply biochemical tools to investigate the relationship between proteins required for vertebrate NHEJ. We show that in vitro end joining catalyzed by the L4·X4 complex occurs by two biochemically distinct pathways. One is the Mre11-Rad50-NBS1-dependent pathway originally defined in yeast, the other involves a novel factor. The two pathways differ with respect to their requirement for DNA-PKcs-mediated phosphorylation, but converge on a common requirement for the L4·X4 complex.
| EXPERIMENTAL PROCEDURES |
|---|
|
|
|---|
DNA End JoiningAssays were performed using 0.5 ng/µl radiolabeled linearized plasmid DNA with product analysis by SDS-agarose gel electrophoresis (17). Gels were 0.6% agarose unless otherwise indicated.
| RESULTS |
|---|
|
|
|---|
|
Previous work has shown that the ligase stimulatory activity of the 0.85 M fraction is attributable, in part, to the presence of the Mre11-Rad50-NBS1 (M·R·N) complex (17). Immunoblotting confirmed the presence of the M·R·N complex in the 0.5 and 0.85 M KOAc fractions. However, this complex, as well as its individual components, were absent from the 0.2 and 0.3 M fractions (Fig. 1D). Stimulatory activity of the 0.2 and 0.3 M fractions must therefore be attributable to a different factor or factors.
Immunoblotting was performed to detect other known DSB repair proteins in the Q-Sepharose fractions. We probed with antibodies to the Werner syndrome gene product (WRN), which interacts with Ku in vitro (18, 19), Artemis, which is required for NHEJ of hairpin-ended DNAs (20), and 53BP1, HDAC4, and MDC1-NFBD1, which localize to DSB sites in vivo (2126). We also probed with antibodies to the SMC1 and Rad21 subunits of cohesin, a protein implicated in DSB repair (27), and to Rad51, a protein required for homologous recombination (28). Except for traces of Rad51 in the 0.3 M fraction, none of these proteins was detectable in the 0.2 M-0.3 M region of the Q-Sepharose column profile. All of these proteins were detectable in unfractionated nuclear extract (Fig. 1D). Results suggest that activity of the 0.20.3 M fractions is not attributable to known NHEJ proteins.
Differential Regulation by DNA-PKcs PhosphorylationOne of the goals of our study was to investigate the regulation of end joining by DNA-PKcs. DNA-PKcs phosphorylation dependence is a distinguishing feature that differentiates vertebrate NHEJ from the simpler process in S. cerevisiae and other model organisms. End joining was performed with various fractions in the presence and absence of a DNA-PKcs inhibitor, LY294002. End joining with the heparin-agarose 0.4 M KOAc fraction was partially inhibited by LY294002 (Fig. 2, lanes 12 and 13). The activities that eluted in different regions of the Q-Sepharose profile, however, are oppositely regulated by LY294002. With only minimal levels of endogenous DNA-PKcs present, LY294002 partially inhibited activity of the 0.3 M fraction (lanes 4 and 5), and actually stimulated activity of the 0.85 M fraction (lanes 8 and 9). With addition of exogenous DNA-PKcs, inhibition of the 0.3 M fraction was more pronounced (lanes 6 and 7), whereas the effect on the 0.85 M fraction was unchanged (lanes 10 and 11). Results indicate that the new pathway, defined by the 0.20.3 M fractions, is significantly dependent on phosphorylation. In contrast, the M·R·N-dependent activity is not phosphorylation-dependent, consistent with previous results (17). The ability of LY294002 to stimulate activity in the presence of the 0.85 M fraction, which was somewhat variable between experiments, could reflect suppression of a competing, phosphorylation-dependent pathway.
|
Further Purification of the Novel ActivityBecause the 0.3 M fraction was contaminated with traces of DNA-PKcs and may have contained other interfering proteins, it was further purified by size exclusion chromatography. A peak of stimulatory activity eluted at a position corresponding to
200 kDa (Fig. 3), well separated from the void volume. The products in this experiment were primarily dimer and trimer products, although in other preparations, where the factor was more concentrated, higher order multimers were also present (see Fig. 5).
|
|
The position of elution of activity on the gel filtration column effectively eliminates the possibility of cross-contamination with M·R·N or DNA-PKcs at levels below the threshold detectable by immunoblotting, as both proteins elute at or near the void volume under the conditions used (17, 29). Results also rule out the involvement of other large complexes, such as BASC, a supercomplex of BRCA1-associated proteins involved in DNA repair (30). We were thus able to evaluate the requirement for the physical presence of DNA-PKcs separately from the requirement for DNA-PKcs-mediated phosphorylation.
DNA-PKcs Establishes a Reaction CheckpointThe novel end joining factor was seen in the absence of DNA-PKcs and this activity was unaffected by LY294002 (Fig. 4A, lanes 6 and 7). DNA-PKcs was separately purified from HeLa cell extracts (Fig. 1A) and added to the reconstituted reactions. There was a small decline in basal activity, and reactions became strongly sensitive to LY294002 inhibition (lanes 912). Thus, physical presence of DNA-PKcs is not required for end joining, but when present, DNA-PKcs restricts the reaction, such that phosphorylation is required for further progression. We term this restriction a "reaction checkpoint" because it represents a point in the reaction pathway where traffic can potentially be halted or diverted (see "Discussion"). This experiment also shows that end joining activity with the more purified fraction is completely dependent on Ku and L4·X4 (lanes 3 and 4), thus eliminating any potential concerns about co-purification of endogenous ligases with the stimulatory fraction. A similar end joining experiment performed in the absence and presence of wortmannin, a different DNA-PKcs inhibitor, confirmed that phosphorylation activity is partially required whenever DNA-PKcs is physically present in the reaction (Fig. 4B).
|
Mechanism of Action of the Novel Stimulatory FactorPrevious work has shown that human Rad50-Mre11 directly tethers DNA ends to each other in vitro (31). This tethering provides a potential mechanism for the stimulatory effect of the M·R·N complex on L4·X4-catalyzed DNA ligation. It was of interest to determine whether the novel 200-kDa repair factor shared a similar mechanism of action. If the factor interacted directly with DNA, one prediction is that it might stimulate the activity of other ligases, in addition to L4·X4.
Fig. 5 compares the effect of the 200-kDa repair factor on end joining catalyzed by L4·X4 versus a homologous ATP-dependent DNA ligase encoded by bacteriophage T4. Consistent with previous experiments, the factor stimulated L4·X4 activity more than 10-fold (compare lanes 1 and 2). The effect of the factor on T4 ligase activity was more modest (a 24-fold stimulation). However, there was a striking redistribution of the products in the T4 ligase reaction. In the absence of the factor, T4 ligase produced a preponderance of circular intramolecular ligation products, as expected at the low concentration of substrate DNA used in the reaction. In the presence of the factor, T4 ligase produced a ladder of concatenated product more typical of the eukaryotic end joining reaction. Thus, the mechanism of action of the factor appears to involve, in part, direct interaction with DNA, in addition to or instead of specific interaction with L4·X4. The effect of this interaction with DNA is to favor intermolecular over intramolecular ligation.
| DISCUSSION |
|---|
|
|
|---|
Surprisingly, we found that phosphorylation was required if and only if DNA-PKcs was physically present in the reaction. DNA-PKcs appears to arrest the reaction until phosphorylation occurs, establishing what may be termed a reaction checkpoint. The checkpoint is released upon receipt of some internally generated signal. Perhaps this signal is generated by the synapsis of opposing DNA ends, which triggers activation of DNA-PKcs in vitro (32). Activation of DNA-PKcs could release the complex from its arrested state through phosphorylation of DNA-PKcs itself, XRCC4, or other targets within the repair complex (33, 34).
A more subtle issue is why a phosphorylation-dependent checkpoint should occur at all in the NHEJ pathway. The existence of a checkpoint implies the possibility of alternate fates. It may be that a delay in phosphorylation, or phosphorylation of different targets within the repair complex, can divert the complex to a different DSB repair pathway. In this respect, it has been shown that the structure of DNA ends can influence the protein substrate specificity of DNA-PKcs (35). It will be of interest to learn if ends that require processing before joining influence DNA-PKcs substrate specificity in such a way as to activate accessory factors required for that processing (such as Artemis, required for hairpin ends) or to divert repair to the M·R·N-dependent or homologous recombination pathways.
The novel 200-kDa repair factor produces a striking change in its ability to alter the distribution of intermolecular versus intramolecular products as seen with T4 ligase. In general, the rate of formation of intermolecular ligation products is directly proportional to molar DNA end concentration. In contrast, the rate of formation of intramolecular products is determined by polymer chain statistics, and in the range of interest declines with increasing DNA length (36). It may be that the factor accelerates intermolecular DNA end joining by increasing the effective DNA end concentration. This could occur if the factor tethered DNA ends to each other, like the Mre11-Rad50 complex. An alternate possibility is that the factor alters the ratio of different products by inhibiting the intramolecular reaction. This could occur if binding of the factor stiffens the DNA chain, altering its conformation, and causing it to behave as if its length were greater.
The active polypeptides in the novel 200-kDa fraction have yet to be identified. Because the factor is not yet pure, we cannot rule out the possibility that more than one component contributes to its activity. In particular, we cannot be certain that the component responsible for redistribution of T4 ligase products is the only factor required for stimulation of L4·X4 activity. Although the identification of these active component(s) will be required to demonstrate relevance of the factor to DSB repair in vivo, the ability of the factor to stimulate L4·X4, the dependence on Ku, and the regulation by DNA-PKcs all argue that the factor is likely to be physiologically relevant.
One significant difference between the in vivo and in vitro results remains, however. In vivo, DNA-PKcs not only regulates end joining but also increases its efficiency. Physical absence of DNA-PKcs, as seen in null mutants, leads to a deficiency in DSB repair. Specifically, such mutants are impaired in the ability to rejoin radiation-induced DSBs and to form V(D)J coding joints, although they remain competent to form V(D)J signal joints and to repair etoposide-induced DSBs (13, 37). In contrast, physical absence of DNA-PKcs in the simplified in vitro system did not affect overall end joining efficiency. It may be that DNA-PKcs increases efficiency in vivo by recruiting processing enzymes required to join certain types of breaks, or by antagonizing negative regulators that are absent from the purified system. Because cohesive 5'-phosphoryl ends, used here, require no processing, and potential negative regulators have been removed by chromatographic fractionation, these functions might become dispensable in vitro. Data presented here may explain the phenotype of DNA-PKcs-deficient mice. Coding joint formation may require DNA-PK kinase activity to process the hairpin coding ends (probably with Artemis), whereas signal joint formation involves simple blunt-end ligation, for which DNA-PKcs may be dispensable.
| FOOTNOTES |
|---|
Both authors contributed equally to this work. ![]()
Eminent Scholar of the Georgia Research Alliance. To whom correspondence should be addressed. Tel.: 706-721-8756; Fax: 706-721-8752; E-mail: wdynan{at}mcg.edu.
1 The abbreviations used are: DSB, double-strand break; NHEJ, nonhomologous end joining; DNA-PKcs, DNA-dependent protein kinase catalytic subunit; WRN, Werner syndrome gene product. ![]()
| ACKNOWLEDGMENTS |
|---|
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
X. Cui and K. Meek Linking double-stranded DNA breaks to the recombination activating gene complex directs repair to the nonhomologous end-joining pathway PNAS, October 23, 2007; 104(43): 17046 - 17051. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Wang, B. Rosidi, R. Perrault, M. Wang, L. Zhang, F. Windhofer, and G. Iliakis DNA Ligase III as a Candidate Component of Backup Pathways of Nonhomologous End Joining Cancer Res., May 15, 2005; 65(10): 4020 - 4030. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Drouet, C. Delteil, J. Lefrancois, P. Concannon, B. Salles, and P. Calsou DNA-dependent Protein Kinase and XRCC4-DNA Ligase IV Mobilization in the Cell in Response to DNA Double Strand Breaks J. Biol. Chem., February 25, 2005; 280(8): 7060 - 7069. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. L. Bladen, D. Udayakumar, Y. Takeda, and W. S. Dynan Identification of the Polypyrimidine Tract Binding Protein-associated Splicing Factor{middle dot}p54(nrb) Complex as a Candidate DNA Double-strand Break Rejoining Factor J. Biol. Chem., February 18, 2005; 280(7): 5205 - 5210. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Kysela, M. Chovanec, and P. A. Jeggo Phosphorylation of linker histones by DNA-dependent protein kinase is required for DNA ligase IV-dependent ligation in the presence of histone H1 PNAS, February 8, 2005; 102(6): 1877 - 1882. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Convery, E. K. Shin, Q. Ding, W. Wang, P. Douglas, L. S. Davis, J. A. Nickoloff, S. P. Lees-Miller, and K. Meek Inhibition of homologous recombination by variants of the catalytic subunit of the DNA-dependent protein kinase (DNA-PKcs) PNAS, February 1, 2005; 102(5): 1345 - 1350. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. J. Romeijn, M. M. Gorski, M. A. van Schie, J. N. Noordermeer, L. H. Mullenders, W. Ferro, and A. Pastink Lig4 and Rad54 Are Required for Repair of DNA Double-Strand Breaks Induced by P-Element Excision in Drosophila Genetics, February 1, 2005; 169(2): 795 - 806. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Jacob, C. Miquel, A. Sarasin, and F.ço. Praz Effects of camptothecin on double-strand break repair by non-homologous end-joining in DNA mismatch repair-deficient human colorectal cancer cell lines Nucleic Acids Res., January 7, 2005; 33(1): 106 - 113. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. V. R. Reddy, Q. Ding, S. P. Lees-Miller, K. Meek, and D. A. Ramsden Non-homologous End Joining Requires That the DNA-PK Complex Undergo an Autophosphorylation-dependent Rearrangement at DNA Ends J. Biol. Chem., September 17, 2004; 279(38): 39408 - 39413. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Weterings, N. S. Verkaik, H. T. Bruggenwirth, J. H. J. Hoeijmakers, and D. C. van Gent The role of DNA dependent protein kinase in synapsis of DNA ends Nucleic Acids Res., December 15, 2003; 31(24): 7238 - 7246. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |