Molecular analysis of Ku redox regulation
© Bennett et al; licensee BioMed Central Ltd. 2009
Received: 11 May 2009
Accepted: 28 August 2009
Published: 28 August 2009
DNA double-strand breaks (DSBs) can occur in response to ionizing radiation (IR), radiomimetic agents and from endogenous DNA-damaging reactive oxygen metabolites. Unrepaired or improperly repaired DSBs are potentially the most lethal form of DNA damage and can result in chromosomal translocations and contribute to the development of cancer. The principal mechanism for the repair of DSBs in humans is non-homologous end-joining (NHEJ). Ku is a key member of the NHEJ pathway and plays an important role in the recognition step when it binds to free DNA termini. Ku then stimulates the assembly and activation of other NHEJ components. DNA binding of Ku is regulated by redox conditions and evidence from our laboratory has demonstrated that Ku undergoes structural changes when oxidized that results in a reduction in DNA binding activity. The C-terminal domain and cysteine 493 of Ku80 were investigated for their contribution to redox regulation of Ku.
We effectively removed the C-terminal domain of Ku80 generating a truncation mutant and co-expressed this variant with wild type Ku70 in an insect cell system to create a Ku70/80ΔC heterodimer. We also generated two single amino acid variants of Cys493, replacing this amino acid with either an alanine (C493A) or a serine (C493S), and over-expressed the variant proteins in SF9 insect cells in complex with wild type Ku70. Neither the truncation nor the amino acid substitutions alters protein expression or stability as determined by SDS-PAGE and Western blot analysis. We show that the C493 mutations do not alter the ability of Ku to bind duplex DNA in vitro under reduced conditions while truncation of the Ku80 C-terminus slightly reduced DNA binding affinity. Diamide oxidation of cysteines was shown to inhibit DNA binding similarly for both the wild-type and all variant proteins. Interestingly, differential DNA binding activity following re-reduction was observed for the Ku70/80ΔC truncation mutant.
Together, these results suggest that the C-terminal domain and C493 of Ku80 play at most a minor role in the redox regulation of Ku, and that other cysteines are likely involved, either alone or in conjunction with these regions of Ku80.
DNA double strand breaks (DSBs) can be caused by ionizing radiation, reactive oxygen species and other endogenous and exogenous events. If these breaks are not repaired they ultimately result in cell death. Inaccurate repair of these breaks can generate chromosomal translocations, deletions and mutations which can lead to genetic instability and contribute to the development and progression of cancer. There are two main pathways to repair DSBs, homologous recombination (HR) and non-homologous end joining (NHEJ). HR occurs with minimal loss of genetic material increasing its accuracy and only occurs when a homologous chromosome is present providing extensive regions of sequence homology. NHEJ is error-prone, however it does not require a homologous chromosome or significant regions of homology and is the predominant pathway to repair IR-induced DNA DSBs. NHEJ is initiated upon Ku binding to the DNA termini generated from the DSB. Subsequent binding of the DNA dependent protein kinase catalytic subunit (DNA-PKcs) forms the activated DNA-PK holoenzyme. Active DNA-PK then catalyzes autophosphorylation and phosphorylation of other downstream NHEJ proteins such as Artemis, MRE11/RAD50/NBS1 (MRN), and DNA ligase IV/XRCC4 .
Ku plays a key role in the NHEJ pathway by binding DNA ends and recruiting other downstream proteins. The crystal structure of Ku revealed a bridge and pillar region comprised of both Ku70 and Ku80 subunits that forms a ring around DNA . These studies revealed the ring shape in the presence and absence of DNA as well as a great deal of structure homology between the two subunits, despite the fact that they share minimal sequence homology. The 3-dimensionial structure of Ku enables the protein to slide or translocate along the length of a DNA molecule. However, it is unclear how Ku dissociates from the DNA upon completion of the NHEJ pathway when the termini are eventually ligated. Additional studies have demonstrated that upon DNA-PKcs binding, Ku translocates inward along the DNA in an ATP independent manner  consistent with the sliding model. Studies have shown that Ku binds DNA in a sequence independent fashion by way of several hydrophobic residues that make contact with the major groove of DNA and several basic residues that interact with the phosphate back bone[6, 8]. Studies have shown that the Ku70 subunit is proximal to the DSB and Ku80 is distal to the DSB.
While much is known about the biochemical activities of Ku, its physiological regulation is less well understood. It has been determined that oxidative stress has a significant effect on the NHEJ pathway. Previous studies have shown that under oxidative conditions there is a marked decrease in DNA-PK activity[9, 10]. More specifically, oxidative stress has been shown to impair Ku's ability to bind DNA. Research has indicated a conformational change in Ku under oxidized conditions that leads to a significantly higher Koff rate . The affect oxidative stress has on Ku is a curious issue when thinking in terms of the crystal structure of Ku. The crystal structure does not reveal any disulfide bonds, however it is lacking several amino acids, particularly a cysteine in the C-terminal region of Ku80.
To further understand how redox conditions influence Ku structure and activity we constructed, purified and characterized several mutants of Ku. These mutations were introduced in key positions of the Ku80 subunit that have been implicated in redox regulation. The results are discussed with respect to the effect of redox on Ku structure and activity.
Sequencing grade bovine trypsin was purchased from Roche Diagnostics (Indianapolis, IN).
DNA primers and oligonucleotides used in this study were purchased from Integrated DNA Technologies, Inc. (Carolville, IA). Srf1 was purchased from Stratagene (La Jolla, CA) and all other restriction enzymes and T4 DNA ligase were purchased from New England Biolabs, Inc. (Ipswich, MA). Mouse monoclonal antibodies Ku (p70) Ab-4 and Ku (p80) Ab-7 were purchased from Neomarker (Fremont, CA).
Single amino acid substitutions of the human Ku80 protein in the pFastBac1 expression vector were generated using the QuikChange II site-directed mutagenesis kit (Stratagene; La Jolla, Ca). Briefly, for each mutation, the plasmid was PCR-amplified using two complementary oligonucleotide primers containing the desired mutation (Table 1). The PCR products were treated with Dpn1 to degrade the methylated parental DNA template. The DNA was then amplified and recombinant baculovirus was generated in the Bac-to-Bac Baculovirus expression system (Invitrogen; Carlsbad, CA). The pFastBac1 expression vector containing the mutated Ku80 gene, was transformed into E. coli strain DH10Bac that contains a baculovirus shuttle vector, bacmid. Transformants were selected and high molecular weight recombinant bacmid DNA was extracted and used for transfection of SF9 cells using FuGENE 6. The clarified transfection supernatant, containing the recombinant baculovirus, was plaque purified as needed and recombinant virus was amplified. Protein expression was accomplished via co-infection with Ku70 virus as previously described [12, 13].
Protein Expression and Purification
Human Ku was purified from Sf9 cells infected with recombinant baculovirus. Cells were infected for 48 hours, and cell-free extracts were prepared. Wild type, Ku70/80ΔC and Ku70/C493 mutants were purified by sequential Ni-NTA and Q-Sepharose column chromatography as previously described [12, 14]. Fractions containing Ku were identified based on SDS-PAGE and visualized by Coomassie Blue staining. Peak fractions were pooled, dialyzed overnight and stored at -80°C
SDS-PAGE and Western Blot
Proteins were separated via SDS-PAGE. Gels were either stained with Coomassie Blue or transferred to PVDF membrane for Western blot analysis. Membranes were blocked with 2% non-fat dry milk in TBS-Tween and probed with the primary antibodies indicated in the figure legends. Bound antibodies were detected with a horse radish peroxidase (HRP) conjugated goat-anti-mouse IgG and visualized via chemiluminescence detection capturing images via a Fuji LAS-3000 CCD system.
Electrophoretic Mobility Shift Assays (EMSAs) were performed as previously described [13, 15]. Briefly, reactions were performed in a volume of 20 μl containing 50 mM Tris-Cl pH7.8, 10 mM MgCl2 and 50 mM NaCl. Oxidized conditions were achieved by incubating Ku for 15 min on ice in 2 mM Diamide. Re-reduced conditions were achieved by incubating oxidized Ku with 5 mM DTT for 15 min on ice. The protein preparations were then assessed for DNA binding activity in an EMSA containing 500 fmol of 32P-labeled 30-bp double strand DNA as previously described using oligonucleotides 30A and 30C (Table 1)[16, 17]. Reaction products were then separated by electrophoresis on a 6% native polyacrylamide gel. The gels were then dried and exposed to a PhosphorImager screen (Amersham Biosciences; Piscataway, NJ) and quantified using ImageQuant software. Quantification of the data is presented as the averages and standard deviations of at least three independent measurements. Binding data were fit to the equation 1 describing a sigmoidal curve and KD values calculated from the fit of the curve using SigmaPlot software (Systat Software inc. Chicago, IL).
Fluorescence polarization experiments were performed in 0.5 ml buffer A (50 mM Tris -HCl, pH 7.8, 10 mM MgCl2s, 50 mM NaCl, 1 mM DTT) using a Cary Eclipse Fluorescence Spectrophotometer (Varian; Palo Alto, CA). Oxidized conditions and re -reduced Ku preparations were generated as described above. The protein preparations were then assessed for DNA binding using a 5'-Fluorscein -labeled 30-bp double-strand DNA as previously described (Table 1)[16, 17]. Fluorescence excitation and emission was measured at 495 and 515 nm, respectively. Results are presented as the averages and standard deviations of at least three independent measurements. KD values were obtained from fitting the data to sigmoidal curves described above. The data obtained from binding under oxidized conditions were not suitable for KD determination.
Limited tryptic proteolysis was performed according to established procedures with the following modification . The Ku preparations were analyzed for potential structural changes under control, reduced, conditions and following oxidation and re-reduction. Ku protein preparations (4 μg), prepared as stated above, were subjected to limited proteolysis by the addition of 200 ng of sequencing grade bovine trypsin. Reactions were performed in buffer A and incubated at 37°C for10 minutes. Reactions were terminated by the addition of SDS loading dye and samples were separated by SDS-PAGE. Products were visualized via Coomassie Blue staining and images were captured using Image Reader LAS-3000 and visualized and quantified using MultiGuage V3.0.
Identification and mutation of potential redox regulated sites in Ku
DNA binding of Ku is independent of the Ku 80 CTD and C493
Redox effects on DNA Binding
To assess the effect of redox on Ku binding we used the cysteine specific oxidant, diamide, to oxidize Ku and then assessed binding in are action performed in the absence of added DTT. Previously we have shown that under these conditions Ku exhibits a reversible oxidation event that impairs DNA binding . The results presented in Figure 3 demonstrate that wild type Ku displays a decrease in DNA binding activity as a function of diamide concentration, consistent with our previous results.
KD Values as determined by EMSA and anisotropy
Wild type Ku
29.7 ± 2.5
51.0 ± 7.7
Wild type Ku
25.8 ± 2.3
31.4 ± 3.0
25.7 ± 1.5
It has been previously shown that the Ku-DNA interaction is favored under reduced conditions[11, 12, 20]. Physiologically, Ku, DNA-PK and NHEJ have all been demonstrated to be influenced by cellular redox conditions[9, 10, 21]. Typically, redox dependent alterations in enzymatic or binding activity can be attributed to the formation and breaking of disulfide bonds. Interestingly, X-ray structural analysis of Ku does not reveal the presence of any disulfide bonds . Therefore, the oxidation effects observed in vitro are likely to be a result of the formation of cysteine sulfenic acids. This reversible modification is consistent with the near complete rescuing of DNA binding activity upon re-reduction (Figure 3 and ref ). Cellular modifications as a function of redox however, may be manifested by different modifications and interactions.
Specifically, glutathionine conjugations to Ku could be responsible for the decreased binding activity observed in cells exposed to ROS via treatment with glucose and glucose oxidase. Also, protein-protein disulfide bonds between exposed cysteine residues could account for the reduction in binding. This, however, is unlikely to account for all the reduction as the Ku protein level was reduced and not observed as highly cross linked . Preliminary data demonstrate that at least 7 cysteine residues are readily accessible, one being located in the C-terminal domain, and thus glutathionine conjugates or modification to sulfenic acid could account for the reduction in DNA binding activity. While the results presented demonstrate that C493 and C638 are not likely to be the major determinants of redox regulation, the inability to completely re-reduce Ku70/80ΔC suggested that C638 may play some role in redox dependent changes in Ku structure.
The DNA binding assays performed on the C493 mutants and Ku70/80ΔC under oxidized conditions revealed a near complete loss of DNA binding activity. If C493 and 638 were responsible for the redox regulation, no reduction in DNA binding activity would have been observed. Demonstration of the loss of binding in two independent assays provides strong evidence that upon oxidation, the protein structure is altered such that it cannot support DNA binding. When the proteins were re -reduced, the DNA binding ability was largely recovered, again consistent with a modification or alteration in structure being reversible. Also, in that the wild type Ku and Ku 70/80ΔC behaved similarly in their recovery suggest that the Ku80CTD does not protect any cysteine residues from the redox conditions that affect Ku's ability to bind DNA. While we observed some discrepancies between the wild type Ku and Ku70/80ΔC anisotropy data and EMSA data, this could be explained in the nature of these assays. The anisotropy assay is more representative of a true steady state equilibrium binding assay, where as the EMSA is a stopped assay with post -binding separation that can influence the detection of the bound species. This is apparent from the fits of the binding curves for each assay where the EMSA quantification yields a clear sigmoidal binding curve as we have previously demonstrated for Ku binding short DNA duplexes . The data obtained from fluorescence anisotropy fits better to a hyperbolic binding curve than the EMSA data, consistent with a true equilibrium binding reaction (data not shown). Despite these differences, the results of the redox effects between the wild type and variant proteins are very consistent.
In conclusion we observed that under oxidized conditions Ku, independent of the mutations, binds DNA with a significantly lower affinity than under reduced conditions. We also determined the C493 does not play a role in Ku-DNA interaction nor does it have a role in the affect of redox on the Ku heterodimer. Analysis of the Ku70/80ΔC revealed slight differences in activity leading us to believe that this truncation has a modest effect on Ku binding DNA and potentially a similar effect on Ku structure and activity in response to redox conditions.
We would like to thank Dr. Dale Ramsden for supplying the Ku plasmid from which all mutants were cloned. We would also like to thank the members of the Turchi lab for critical reading of this manuscript as well as stimulating conversations. This work was supported by grant CA82741 from the National Institutes of Health to JJT. TMN was supported by NIH Cancer Biology training grant T32CA111198
- Delacote F, Guirouilh-Barbat J, Lambert S, Lopez BS: Homologous recombination, non-homologous end-joining and cell cycle: Genome's angels. Current Genomics. 2004, 5: 49-58. 10.2174/1389202043490041..View ArticleGoogle Scholar
- Yoo S, Dynan WS: Geometry of a complex formed by double strand break repair proteins at a single DNA end: recruitment of DNA-PKcs induces inward translocation of Ku protein. Nucleic Acids Res. 1999, 27: 4679-4686. 10.1093/nar/27.24.4679PubMed CentralView ArticlePubMedGoogle Scholar
- Poinsignon C, de Chasseval R, Soubeyrand S, Moshous D, Fischer A, Hache RJG, et al: Phosphorylation of Artemis following irradiation-induced DNA damage. European Journal of Immunology. 2004, 34: 3146-3155. 10.1002/eji.200425455View ArticlePubMedGoogle Scholar
- Uziel T, Lerenthal Y, Moyal L, Andegeko Y, Mittelman L, Shiloh Y: Requirement of the MRN complex for ATM activation by DNA damage. EMBO J. 2003, 22: 5612-5621. 10.1093/emboj/cdg541PubMed CentralView ArticlePubMedGoogle Scholar
- Wang HC, Zeng ZC, Perrault AR, Cheng XB, Qin W, Iliakis G: Genetic evidence for the involvement of DNA ligase IV in the DNA-PK-dependent pathway of non-homologous end joining in mammalian cells. Nucleic Acids Res. 2001, 29: 1653-1660. 10.1093/nar/29.8.1653PubMed CentralView ArticlePubMedGoogle Scholar
- Walker JR, Corpina RA, Goldberg J: Structure of the Ku heterodimer bound to DNA and its implications for double-strand break repair. Nature. 2001, 412: 607-614. 10.1038/35088000View ArticlePubMedGoogle Scholar
- Turchi JJ, Henkels KM, Zhou Y: Cisplatin-DNA adducts inhibit translocation of the Ku subunits of DNA-PK. Nucleic Acids Res. 2000, 28: 4634-4641. 10.1093/nar/28.23.4634PubMed CentralView ArticlePubMedGoogle Scholar
- de Vries EG, van Driel W, Bergsma WG, Arnberg AC, Vliet van der PC: HeLa nuclear protein recognizing DNA termini and translocating on DNA forming a regular DNA-multimeric protein complex. J Mo Biol. 1989, 208: 65-78. 10.1016/0022-2836(89)90088-0..View ArticleGoogle Scholar
- Boldogh I, Roy G, Lee MS, Bacsi A, Hazra TK, Bhakat KK, et al: Reduced DNA double strand breaks in chlorambucil resistant cells are related to high DNA-PKcs activity and low oxidative stress. Toxicology. 2003, 193: 137-152. 10.1016/j.tox.2003.08.013View ArticlePubMedGoogle Scholar
- Bacsi A, Kannan S, Lee MS, Hazra TK, Boldogh I: Modulation of DNA-dependent protein kinase activity in chlorambucil-treated cells. Free Radical Biology and Medicine. 2005, 39: 1650-1659. 10.1016/j.freeradbiomed.2005.08.013View ArticlePubMedGoogle Scholar
- Zhang WW, Yaneva M: Reduced sulphydryl groups are required for DNA binding of Ku protein. Biochemical Journal. 1993, 293: 769-774.PubMed CentralView ArticlePubMedGoogle Scholar
- Andrews BJ, Lehman JA, Turchi JJ: Kinetic analysis of the Ku-DNA binding activity reveals a redox-dependent alteration in protein structure that stimulates dissociation of the Ku-DNA complex. J Biol Chem. 2006, 281: 13596-13603. 10.1074/jbc.M512787200PubMed CentralView ArticlePubMedGoogle Scholar
- Lehman JA, Hoelz DJ, Turchi JJ: DNA-Dependent Conformational Changes in the Ku Heterodimer. Biochemistry. 2008, 47: 4359-4368. 10.1021/bi702284cPubMed CentralView ArticlePubMedGoogle Scholar
- McElhinny SAN, Snowden CM, McCarville J, Ramsden DA: Ku recruits the XRCC4-ligase IV complex to DNA ends. Mol Cell Biol. 2000, 20: 2996-3003. 10.1128/MCB.20.9.2996-3003.2000View ArticleGoogle Scholar
- Turchi JJ, Henkels KM: Human Ku autoantigen binds cisplatin-damaged DNA but fails to stimulate human DNA-activated protein kinase. J Biol Chem. 1996, 271: 13861-13867. 10.1074/jbc.271.23.13861View ArticlePubMedGoogle Scholar
- Hermanson IL, Turchi JJ: Overexpression and purification of human XPA using a Baculovirus expression system. Protein Expression and Purification. 2000, 19: 1-11. 10.1006/prep.2000.1224View ArticlePubMedGoogle Scholar
- Hermanson-Miller IL, Turchi JJ: Strand-specific binding of RPA and XPA to damaged duplex DNA. Biochemistry. 2002, 41: 2402-2408. 10.1021/bi0112863View ArticlePubMedGoogle Scholar
- Turchi JJ, Patrick SM, Henkels KM: Mechanism of DNA-dependent protein kinase inhibition by cis- diamminedichloroplatinum(II)-damaged DNA. Biochemistry. 1997, 36: 7586-7593. 10.1021/bi963124qView ArticlePubMedGoogle Scholar
- Singleton BK, Torres-Arzayus MI, Rottinghaus ST, Taccioli GE, Jeggo PA: The C terminus of Ku80 activates the DNA-dependent protein kinase catalytic subunit. Mol Cell Biol. 1999, 19: 3267-3277.PubMed CentralView ArticlePubMedGoogle Scholar
- Ayene IS, Stamato TD, Mauldin SK, Biaglow JE, Tuttle SW, Jenkins SF, et al: Mutation in the glucose-6-phosphate dehydrogenase gene leads to inactivation of Ku DNA end binding during oxidative stress. J Biol Chem. 2002, 277: 9929-9935. 10.1074/jbc.M111366200View ArticlePubMedGoogle Scholar
- Song JY, Lim JW, Kim H, Morio T, Kim KH: Oxidative stress induces nuclear loss of DNA repair proteins Ku70 and Ku80 and apoptosis in pancreatic acinar AR42J cells. J Biol Chem. 2003, 278: 36676-36687. 10.1074/jbc.M303692200View ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.