Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

DNA-dependent protein kinase (DNA-PK) is a nuclear serine/threonine protein kinase composed of 460-kDa catalytic subunit (DNA-PKcs) and a heterodimer of Ku70 and Ku80, which act as a DNA binding and regulatory component for the complex (Gottlieb and Jackson, 1993; Jin et al., 1997; Lieber et al., 1997). DNA-PK is a key component of the nonhomologous end joining (NHEJ) pathway and V(D)J recombination (Blunt et al., 1995), with the unique property of being activated by DNA ends (Critchlow and Jackson, 1998; Jeggo, 1998; Featherstone and Jackson, 1999). It has long been suspected as a factor involved in sensing and transmitting DNA damage signals to the downstream targets (Jackson, 1997; Weaver and Alt, 1997; Lee and Kim, 2002). Previous studies demonstrated that DNA-PK is necessary for activation of p53 (Woo et al., 1998), nucleotide excision repair (Muller et al., 1998), and damage-induced S-phase arrest (Park et al., 1999) in response to DNA damage, all of which contribute to cell protection from genetic alterations as well as chemotherapy drug resistance. In vivo observations indicated that DNA-PK mutant cells exhibited sensitivity to ionizing irradiation and chemotherapy drugs and were associated with lower DNA repair activity following DNA damage, suggesting a positive role for DNA-PKcs in DNA repair (Britten et al., 1999; Frit et al., 1999). Also, studies with drug-resistant or drug-sensitive cancer cells suggested that higher levels of DNA-PK expression lead to drug-resistant cells, whereas low DNA-PK activity was associated with cells with drug-sensitive phenotype (Shen et al., 1997) and was linked to cell death via the accumulation of damaged DNA.

The current model of DNA-PK complex activation by DNA is based on the tenet that without DNA, DNA-PKcs is inactive and incapable of binding Ku (Hanawalt, 1994; Suwa et al., 1994; Hartley et al., 1995). When a double-strand break is introduced, Ku complex binds to the DNA because of its high affinity for DNA ends. The binding of Ku induces conformational change that allows it to interact with DNA-PKcs. It is unclear how the Ku/DNA complex activates the kinase activity of DNA-PKcs. One hypothesis is that DNA-PKcs undergoes a conformational change upon association with the Ku/DNA complex, and this conformational change accounts for the activation of kinase activity. The kinase activity associated with DNA-PK is needed for DNA repair in vivo, since expression of a kinase-inactive form of DNA-PKcs failed to complement the radiosensitive phenotype of a mammalian cell line lacking the DNA-PKcs protein (Kurimasa et al., 1999). However, the physiological targets of DNA-PK in vivo are still not clear. The DNA-PK complex can physically tether two ends of a dsb in close proximity in vitro, suggesting the hypothesis that the DNA-PK complex acts as a scaffold to assemble the NHEJ pathway proteins at a DSB (Cary et al., 1997).

We hypothesize that 1) DNA-PK plays an important role in conferring cells becoming resistant to ionizing radiation or anticancer DNA-damaging drugs, and 2) targeted inhibition of DNA-PK sensitizes drug resistance of cancer cells and facilitates cell killing. By developing peptides that can directly interfere with DNA-PK activity, one can develop a novel cotherapy that can selectively target and disrupt the IR-induced dsb repair pathway, which will enhance the efficacy of currently available treatments and also broaden the usefulness of chemotherapeutic agents in cancer treatment. We have therefore synthesized a peptide (HNI-38) mimicking the domain of Ku80 essential for interaction with its catalytic subunit (DNA-PKcs) and tested whether it can selectively target and disrupt DNA-PK activity required for dsb repair, which potentiates the effect of chemotherapy drug in cancer treatment. This strategy can be applied to cancer cotherapy, which will broaden the usefulness of chemotherapeutic agents in cancer treatment.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Lines, Antibodies, and Chemicals. Two human breast cancer cells, MDA231 and NCI, were obtained from Dr. George Sledge (Indiana University Cancer Center, Indianapolis, IN) and maintained in minimal essential medium supplemented with 10% fetal bovine serum at 37°C in a CO₂ incubator. Antibodies to Ku70/80 and the catalytic subunit of DNA-dependent protein kinase (DNA-PKcs) were obtained from either Upstate Biotechnology (Lake Placid, NY) or BD PharMingen (San Diego, CA). [-³²P]ATP (4500 Ci/mmol) was from ICN Pharmaceuticals (Costa Mesa, CA), and dsDNA cellulose and cisplatin were obtained from Sigma-Aldrich (St. Louis, MO).

dsDNA Cellulose Pull-Down Assay. The dsDNA cellulose fraction (100 µg) containing DNA-PKcs and Ku70/Ku80 heterodimer was prepared from HeLa cells (Lees-Miller et al., 1990) and incubated with the indicated amount of either control (HN-26) or a target (HNI-38) peptide in the presence of 4 mM ATP and 50 µl of dsDNA cellulose (3 mg of dsDNA/mg of cellulose; Sigma-Aldrich) for 3 h at 4°C with rocking for the interaction of DNA-PKcs and Ku70/Ku80 heterodimer. Where indicated, purified Ku70/Ku80 complex (100 ng) was used instead of the dsDNA cellulose fraction. After centrifugation at 4000 rpm, the precipitates were collected and washed three times with a buffer (50 mM Tris-HCl, pH 8.0, 200 mM NaCl, 0.5% Nonidet P-40, 1 mM EDTA, 10% glycerol, 1 mM dithiothreitol, and 0.1 mM phenylmethylsulfonyl fluoride) for protein analysis. For Western blot, the precipitates were separated by 8 or 10% SDS-PAGE, transferred to nitrocellulose (Millipore Corp., Bedford, MA), and blotted with primary antibody to Ku70/80 and/or DNA-PKcs followed by a peroxidase-coupled secondary antibody (Amersham Biosciences Inc.,, Piscataway, NJ) and an enhanced chemiluminescence (ECL kit, Amersham) reaction prior to visualization on Kodak-O-mat film.

DNA-PK Kinase Assay. Reaction mixtures (20 µl) contained 20 mM HEPES-KOH (pH 7.5), 1 mM EGTA, 1 mM EDTA, 1 mM dithiothreitol, 10 mM MgCl₂, 7 mM MnCl₂, 5 mM NaF, 1 mM Na₃VO₄, 50 µl of [³²P]ATP, 150 µl of substrate peptide, 0.4 µg of DNase I-activated calf-thymus DNA (Sigma-Aldrich), and 100 ng of partially purified DNA-PK complex. DNA-PK complex was partially purified from HeLa cells according to the procedure described previously (Lees-Miller et al., 1990). Substrate peptide (EPPLSQEAFADLWKK) representing amino acids 11-24 of p53 was used as a substrate for DNA-PK assay (Lees-Miller et al., 1990). To find out whether the peptide interferes with DNA-PK kinase activity, various amounts of peptide inhibitor were added to the reaction. After incubation at 30°C for 30 min, the reaction mixtures were stopped with 30% acetic acid and a portion of the reaction mixtures (5 µl) was spotted onto a P81 strip, and after extensive washing, radioactivity was measured. DNA-PK activity was measured as picomoles of ³²P transferred to the substrate peptide.

Cell Survival Assay. Cells (1.0 × 10⁴ cells/well) were seeded in a 96-well plate in the presence of control or target peptide and incubated for 24 h before the treatment of cells with either ionizing radiation or cisplatin. After further incubation at 37°C, 5% CO₂ for 72 h, cell survival was measured using a colorimetric cell survival assay from Roche Diagnostics (Indianapolis, IN; MTT Cell Proliferation Kit). Alternatively, clonogenic assay was used to measure the ability of cells to form colonies on 100-mm² tissue culture dishes after treatment with ionizing radiation or cisplatin. Controls consisted of cells untreated with peptides, cells untreated with DNA-damaging agent, or cells without either treatment. Cells were continuously exposed for 5 days to the indicated concentrations of the peptide and colonies were stained with crystal violet; then, colonies greater than 50 cells were counted. Each point represents mean values ± S.E., each conducted with triplicate plates. The p values in Fig. 5A (see under Results) were obtained from two separate experiments using a one-way analysis of variance method (SigmaStat for Windows, version 2.03; SPSS Science, Chicago, IL).

Double-Stranded DNA Break (dsb) Repair Assay. Kinetics of rejoining of radiation-induced damaged DNA in breast cancer cells following exposure of cells to 40 Gy gamma irradiation (¹³⁷Cs) were measured by pulsed field gel electrophoresis. Breast cancer cells (NCI) were grown in the presence of 2.5 µM [¹⁴C]thymidine (0.1 µCi/ml) (DiBiase et al., 2000) and treated with either a control or target peptide. After irradiation (40 Gy), cells were further incubated at 37°C with prewarmed (42°C) fresh medium to allow DSB repair, and then harvested at various times and resuspended in serum-free medium at a concentration of 2 to 5 × 10⁶ cells/ml. Cells were mixed with an equal volume of 1% agarose, and the solidified cell-agarose suspension was lysed with buffer containing 10 mM Tris (pH 8.0), 50 mM NaCl, 0.5 M EDTA, 2% N-lauryl sarcosyl, and proteinase E and O (0.1 mg/ml) for 16 to 18 h at 50°C (DiBiase et al., 2000). DNA double-strand breaks were analyzed by asymmetric field inversion gel electrophoresis using 0.5% agarose gel in 0.5× Tris borate-EDTA at 10°C for 40 h. After electrophoresis, gels were analyzed by fluorography. For quantification of damaged DNA repair, intact chromosome and damaged DNA were separately removed from the gel and measured for ¹⁴C labeling using a liquid scintillation counter.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Targeted Inhibition of DNA-PK. Ku70 and Ku80 form a heterodimeric complex that is important for DNA-termini binding; neither Ku70 nor Ku80 alone is active in DNA binding activity (Wu and Lieber, 1996; Gell and Jackson, 1999). The C termini of both Ku70 and Ku80 are necessary for heterodimer assembly as well as for DNA-termini binding (Wu and Lieber, 1996; Gell and Jackson, 1999). A recent protein interaction study indicated that the DNA-PKcs interacting domain is localized at the extreme C terminus of Ku80 (amino acids 720-732) (Gell and Jackson, 1999). Since the C terminus of Ku80 is also likely involved in heterodimer assembly and DNA-termini binding, this region (amino acids 720-732 of Ku80) was selected to synthesize a target peptide that would prevent DNA-PKcs from binding to Ku70/Ku80 regulatory subunits (see Fig. 1A). To deliver a peptide to the cancer cells, a cell-permeable peptide import domain and the nuclear localization domain were added to the target peptide to obviate the need for permeabilization or microinjection of individual cells (Lin et al., 1995; Fig. 1B).

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Panel A, a peptide cotherapy strategy for targeted inhibition of DNA-PK in cancer cell cotherapy. Treatment of cells with ionizing radiation (or chemotherapy drug) induces strand-break DNA damage. To repair DNA damage, DNA-PK heterotrimeric complex (Ku70, Ku80, and DNA-PKcs) needs to be assembled at the ends of DNA. Target peptide representing amino acids 720-732 of Ku80 not only interacts with DNA-PKcs, but may also be involved in Ku heterodimer assembly and DNA-termini binding. As a result, cells treated with target peptide will exhibit poor or no DNA repair and become highly sensitive to ionizing radiation or chemotherapy drug. Panel B, synthetic peptide used for cotherapy study. Sequences of synthetic peptide-based inhibitor and control peptides (single-letter amino acid code). The membrane-translocating hydrophobic signal sequence is indicated in italic letters and the nuclear localization sequence is shown in boldface. Twelve residues of the peptide inhibitor region are indicated at the C terminus of HNI-38.

A Target Peptide Interrupts the Interaction between DNA-PKcs and Ku70/Ku80 as well as the Binding of Ku Complex to DNA. DNA-PKcs and Ku70/Ku80 are abundant proteins, approximately 5 × 10⁵ molecules per human cell (Lee and Kim, 2002, and references therein), and most of the Ku70/Ku80 heterodimer exists in cell extracts without forming a complex with DNA-PKcs in the absence of DNA (Hammarsten and Chu, 1998; Fig. 2A). Therefore, target peptide (HNI-38) was analyzed for its effect on interaction between DNA-PKcs and Ku70/Ku80 in the presence of dsDNA. Varying concentrations of either control (HN-26) or target peptide (HNI-38) were incubated with cell extracts containing DNA-PKcs and Ku complex in the presence of dsDNA cellulose, and examined for its effect on binding of Ku complex and DNA-PKcs to DNA after the dsDNA cellulose pull-down assay (Fig. 2B). Although it was marginal, the addition of increasing amounts of target peptide (HNI-38), not control peptide (HN-26), led to a decrease in DNA-PKcs associated with dsDNA, suggesting that target peptide binds to DNA-PKcs and inhibits its binding to Ku70/Ku80. It is also noted that the addition of target peptide affected the binding of Ku70/Ku80 to the dsDNA cellulose (Fig. 2B). To further examine the effect of HNI-38 on the DNA binding activity of Ku, target peptide was incubated with purified Ku70/Ku80 complex in the presence of dsDNA cellulose, and the reaction mixtures were analyzed for the presence of Ku70 and Ku80 after the dsDNA pull-down assay (Fig. 2C). In keeping with Fig. 2B, target peptide (HNI-38) significantly interfered with binding of Ku complex to dsDNA under the conditions where control peptide (HN-26) showed virtually no effect (Fig. 2C). This result suggests that target peptide not only affects the interaction between DNA-PKcs and Ku complex, but also interferes with the DNA binding activity of Ku.

View larger version (35K): [in this window] [in a new window]   Fig. 2.   Effect of the target peptide on interaction of Ku70/Ku80 with DNA-PK or with dsDNA. Panel A, Ku70/Ku80 complex is not in complex with DNA-PKcs. Chromatographic separation of DNA-PKcs from Ku70/Ku80 heterodimer. Partially purified DNA-PK fractions (dsDNA cellulose fraction; see Lees-Miller et al., 1990 for details) were subjected to heparin-Sepharose column chromatography and eluted with 100 to 500 mM NaCl gradient. Fractions were analyzed by 6% SDS-PAGE for DNA-PKcs and Ku80 followed by immunoblot using anti-DNA-PKcs and anti-Ku80 antibodies. Panel B, the target peptide (HNI-38) interferes with association of DNA-PKcs with dsDNA. Partially purified DNA-PK fraction (100 ng) was incubated with 0 nM (lane 2), 10 nM (lanes 3 and 6), 50 nM (lanes 4 and 7), and 100 nM (lanes 5 and 8) concentrations of either control peptide or target peptide prior to dsDNA cellulose pull-down assay. Lane 1 contained partially purified DNA-PK without dsDNA pull-down assay. The protein-dsDNA cellulose complex was analyzed by the procedure described under Materials and Methods. Panel C, effect of HNI-38 on DNA binding activity of Ku70/Ku80 complex. Purified Ku70/Ku80 complex (100 ng) was incubated with 10 nM (lanes 4 and 7), 50 nM (lanes 3 and 6), and 100 nM (lanes 2 and 5) concentrations of either control peptide or target peptide prior to dsDNA cellulose pull-down assay. After the dsDNA cellulose pull-down, Ku70 and Ku80 were analyzed by 10% SDS-PAGE and Western blot.

Effect of Target Peptide (HNI-38) on DNA-PK Kinase Activity. Interaction of DNA-PKcs with Ku complex is necessary for activation of its kinase activity (Gottlieb and Jackson, 1993; Hartley et al., 1995); therefore, the efficacy of target peptide was analyzed by measuring DNA-PK kinase activity in vitro in the presence of either HN-26 or HNI-38. DNA-PK kinase activity was inhibited up to 50% in the presence of HNI-38 under the conditions where a control peptide (HN-26) showed minimal effect (Fig. 3), strongly supporting the notion that target peptide specifically binds to DNA-PKcs and interferes with interaction between DNA-PKcs and Ku complex. Inhibitory effect of target peptide on DNA-PK occurred at low peptide concentrations (<20 nM) and, in the presence of 20 nM or higher, both target and control peptides inhibited DNA-PK activity (data not shown).

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Effect of target peptide on DNA-PK kinase activity in vitro. Partially purified DNA-PK fraction was incubated with various concentrations of either control peptide or target peptide prior to the addition of substrate peptide and other components for DNA-PK kinase assay (see Materials and Methods for details). DNA-PK activity was measured as the relative amounts of 32P transferred to the substrate peptide.

Target Peptide Interferes with Repair of Double-Stranded DNA Breaks Induced by IR. IR-induced double-stranded DNA breaks are efficiently repaired by an NHEJ process. Genetic and biochemical studies strongly indicated that DNA-PK plays an essential role in NHEJ (Blunt et al., 1995; Jin et al., 1997; Jeggo, 1998). Hence, an alternative way to determine the efficacy of peptide inhibitor is to measure the repair of double-stranded DNA breaks following IR. Breast cancer cells (NCI) grown in the presence of [¹⁴C]thymidine (DiBiase et al., 2000) were treated with either a control or a target peptide for 24 h. After irradiation (40 Gy), cells were harvested at various time points, and intact chromosomal DNA and DSBs were separated by pulsed field gel electrophoresis (0.5% agarose). Treatment of NCI cells with IR (40 Gy) induced substantial amounts of dsDNA breaks, most of which were repaired within 4 h. Cells treated with target or control peptide did not show any difference in generating DSBs after IR (Fig. 4A; lane 2 versus lanes 8 and 14). On the other hand, cells treated with target peptide (Fig. 4A, lanes 14-18) compared with those treated with control peptide (Fig. 4A, lanes 8-12) showed a noticeable decrease in DSB repair activity. This result suggests that target peptide interfered with dsb repair in vivo through the targeted inhibition of DNA-PK activity.

View larger version (39K): [in this window] [in a new window]   Fig. 4.   Effect of target peptide on double-stranded DNA break repair. Breast cancer cells (NCI) grown in 14C-containing media were treated with ionizing radiation (40 Gy) in the presence of no peptide (lanes 1-6), 50 nM control peptide (lanes 7-12), or 50 nM target peptide (lanes 13-18). After harvesting the cells at various time points, intact chromosomes and double-stranded DNA breaks (DSBs) were separated by gel electrophoresis (panel A, fluorography) and were quantified by liquid scintillation counter (panel B).

Target Peptide Inhibits Breast Cancer Cell Growth Only in the Presence of DNA Damage. Cells lacking DNA-PK catalytic subunit showed increased sensitivity to DNA-damaging drugs or IR (Kirchgessner et al., 1995; Lees-Miller et al., 1995), suggesting that DNA-PK activity is essential for DNA repair and cell survival upon DNA damage. We therefore tested whether a targeted inhibition of DNA-PK by a peptide, HNI-38, would sensitize breast cancer cells upon treatment of ionizing radiation or chemotherapeutic drug (cisplatin). Two breast cancer cell lines (NCI and MDA231) were treated with either control (HN-26) or target peptide (HNI-38) and tested for the efficacy of DNA-PK inhibitory peptide on lowering resistance of cells in response to ionizing radiation using a standard colony count cell survival assay. Neither control nor target peptides showed any effect on cell growth in the absence of ionizing radiation. However, cells treated with IR showed significant cell growth inhibition in the presence of target peptide but not with control peptide (Fig. 5A), suggesting that cell growth inhibition by target peptide occurs through targeting DNA-PK activity. Cells treated with cisplatin, although not as effective as those treated with ionizing radiation, also showed inhibitory effect on cell growth in the presence of HNI-38 (Fig. 5B).

View larger version (35K): [in this window] [in a new window]   Fig. 5.   Effect of control (HN-26) and target (HNI-38) peptides on the growth of breast cancer cells treated with ionizing radiation (panel A) or cisplatin (panel B). Values expressed are means (±S.E.) of the three replications (, p < 0.01; , p < 0.01). The clonogenic assay was used for the cells treated with ionizing radiation, and the cell survival assay (MTT) was used for those treated with cisplatin (see Materials and Methods for the detailed procedure).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Many key human DNA repair pathways, such as double-strand break repair or nucleotide excision repair pathway, rely on multimeric polypeptide activities (Friedberg, 1996; Sancar, 1996; Wood, 1996; Lee, 2001). Interactions between damage recognition proteins and those proteins that report the damage to downstream repair activities are crucial for DNA repair. DNA-PK is a key component of the NHEJ pathway with the unique property of being activated by double-stranded DNA breaks (Blunt et al., 1995). Earlier studies with drug-resistant and -sensitive cancer cells suggested that high level expression of DNA-PK leads to drug-resistant cells, whereas low DNA-PK activity was associated with drug-sensitive phenotype (Muller and Salles, 1997; Shen et al., 1997, 1998; Muller et al., 1998; Tew et al., 1998; Frit et al., 1999; Kim et al., 1999, 2000), implicating a role for DNA-PK in conferring cells becoming drug resistant in response to anticancer DNA-damaging drug. Since the interaction of DNA-PKcs to its regulatory subunits, Ku70 and Ku80, is crucial for its function in DNA repair, a targeted inhibition of DNA-PK would sensitize drug resistance of cancer cells and facilitate cell killing. Therefore, we attempted to develop a peptide cotherapy strategy in which a low molecular weight peptide-based inhibitor specifically interferes with interaction between DNA-PKcs and Ku complex.

A target peptide (HNI-38) containing the C terminus of Ku80 interfered with the interaction between DNA-PKcs and Ku complex. This was a much anticipated result since the C terminus of Ku80 was previously identified as DNA-PKcs interacting domain (Gell and Jackson, 1999). Inhibitory effect of HNI-38 on the interaction between DNA-PKcs and Ku70/Ku80 directly affected its kinase activity, showing inhibition of DNA-PK activity up to 50% under the conditions where a control peptide (HN-26) showed very little effect (Fig. 3). However, addition of an excess amount of target peptide did not show any further inhibition of DNA-PK kinase activity (data not shown). This is likely due to the fact that DNA-PKcs without Ku complex can still function as a kinase, although its activity is low. A target peptide (HNI-38) not only inhibited the interaction of DNA-PKcs with Ku complex on dsDNA, but also affected the dsDNA binding activity of Ku (Fig. 2C). It is not clear how HNI-38 interferes with DNA binding activity of the Ku complex; however, the C terminus of both Ku70 and Ku80 has been shown to be important for heterodimer assembly as well as for DNA-termini binding (Wu and Lieber, 1996; Gell and Jackson, 1999). It is possible that HNI-38 may interfere with the Ku70-Ku80 interaction through its binding to Ku70, which would negatively influence the DNA-termini binding activity of Ku.

DNA-PK activity is essential for DNA repair as well as cell cycle arrest in response to DNA damage, which contributes to cell survival by protecting cells from apoptosis. Cells treated with target peptide but not control peptide showed a noticeable decrease in DSB repair after a high dose (40 Gy) of IR, suggesting that HNI-38 specifically targets DNA-PK in vivo and interferes with dsb repair activity through inhibition of DNA-PK activity. Targeted inhibition of DNA-PK by HNI-38 also caused cell growth inhibition only when cells were treated with IR, suggesting that HNI-38 targeted DNA-PK and lowered resistance of cells in response to ionizing radiation, which eventually causes growth inhibition of both NCI and MDA231. Treatment of cells with HNI-38 also showed additive effect on cell growth inhibition in response to cisplatin treatment. This observation is in keeping with previous findings that DNA-PK is directly involved in nucleotide excision repair action in mammals (Muller et al., 1998). It also supports the notion that a targeted inhibition of DNA-PK would sensitize cancer cells upon treatment of chemotherapeutic drugs such as cisplatin. Taken together, our study results described here not only validate DNA-PK as a useful molecular target for the treatment of drug-resistant cancer cells, but also support a physiological role for DNA-PK in IR or chemotherapy drug resistance of cancer cells.