Ku Protein Targeting by Ku70 Small Interfering RNA Enhances Human Cancer Cell Response to Topoisomerase II Inhibitor and ; Radiation

529

Ku protein targeting by Ku70 small interfering RNA enhances human cancer cell response to topoisomerase II inhibitor and ; radiation

Iraimoudi S. Ayene,1 Lance P. Ford,2 Introduction 1 and Cameron J. Koch It has been a long-standing interest to target proteins g 1Department of Radiation Oncology, School of Medicine, responsible for the resistance of cancer cells to radiation University of Pennsylvania, Philadelphia, Pennsylvania and and chemotherapeutic agents (1–4). Most recent molecular 2Department of Research and Development, Ambion Inc., biological studies have focused on the signaling transduc- Austin, Texas tion mechanisms responsible for cellular proliferation and growth (4–7). It is quite possible that targeting signaling Abstract pathways mediated by phosphorylation, kinases, growth Ku protein is a heterodimer (Ku70 and Ku86) known to receptors, and oncogenes may yield clinically relevant play an important role in V(D)J recombination, apoptosis, approaches to specifically kill cancer cells; provided that telomere fusion, and double-strand break repair. Its role in only cancer cells and not the normal cells rely on these double-strand breaks is relevant to cancer therapy be- pathways for survival. There are numerous preclinical cause lack of Ku86 causes one of the most radiation- evidences that these inhibitors can potentiate the antitumor responsive phenotypes (hamster cells, XRS5). Although it effects of many cytotoxic agents (8). However, the efficacy is known that the heterodimer is necessary for the various of these inhibitors to enhance the effects of cytotoxic agents functions of this protein, the impact of targeting Ku in in clinical studies in humans is not clearly established (8). human cancer cells has not been shown due to lack of Baker et al. (9) have shown that overexpression of appropriate approaches. It is also not known whether thioredoxin, which plays some role in the growth of several complete knock-out of Ku protein is required to enhance cancer cells, is effective in reducing the sensitivity of the sensitivity of human cells to ; radiation as Ku protein is mammalian cells to certain chemotherapeutic agents. much more abundant in human cells than in hamster cells. However, we have recently shown that inhibition of the In the current article, we have investigated the direct oxidative pentose phosphate cycle, which supplies effect of Ku70 depletion in human cervical epithelioid NADPH required for the function of thioredoxin and other proteins, did not sensitize the noncancerous Chinese (HeLa) and colon carcinoma (HCT116) cells. We specifi- 3 cally targeted Ku70 mRNA by use of small interfering RNA hamster ovary cells to etoposide, and only slightly (siRNA). Of the five Ku70 siRNA synthesized, three sensitized these cells to g radiation (10). These results have inhibited the expression of Ku70 by up to 70% in HeLa indicated that oxidative pentose phosphate cycle–mediated cells. We have tested the effect of chemically synthesized redox signaling might play only a minor role in cellular siRNAs for target sequence 5 (CS #5) on the response of response to DNA-damaging agents. The most relevant HeLa cells 72 hours after transfection to ; radiation and downstream signaling pathways that could play a signifi- etoposide, as this showed the maximum inhibition of Ku70 cant role in sensitizing the cancer cells to DNA-damaging expression. Ku70 siRNA induced a decrease in the agents are those involved in apoptosis (11–13). We have surviving fraction of irradiated HeLa cells by severalfold. previously shown that DNA is an important target in Similar sensitizing effects were observed for etoposide, a radiation-induced apoptosis, suggesting that both mitotic topoisomerase II inhibitor. Studies with HCT116 cells and apoptotic cell death are caused by DNA lesions after using the same Ku70 siRNA (CS #5) showed a direct irradiation (14). correlation between expression of Ku70 and sensitization Ku protein plays a major role in the repair of DNA to radiation and etoposide treatments. [Mol Cancer Ther double-strand breaks caused by g radiation and some 2005;4(4):529–36] chemotherapeutic agents (15–19). A key system for the repair of DNA double-strand breaks is nonhomologous end-joining (20, 21). The role of Ku is to bind to DNA ends, thus facilitating the coordination of other DNA repair Received 5/21/04; revised 1/20/05; accepted 2/21/05. proteins (20, 21). Biochemical and genetic studies using Grant support: National Cancer Institute research grant CA 92108. mutant rodent cell lines sensitive to ionizing radiation have The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked identified at least four genes, XRCC4, XRCC5, XRCC6, and advertisement in accordance with 18 U.S.C. Section 1734 solely to XRCC7, that are required for nonhomologous end-joining indicate this fact. (20, 21). XRCC4 encodes a 38 kDa nuclear phosphoprotein Requests for reprints: Iraimoudi S. Ayene, Department of Radiation Oncology, School of Medicine, University of Pennsylvania, 195 John Morgan Building, Philadelphia, PA 19104-6072. Phone: 215-898-9507; Fax: 215-898-0090. E-mail: [email protected] Copyright C 2005 American Association for Cancer Research. 3 J.E. Biaglow et al., personal communication.

MolCancerTher2005;4(4).April2005

that binds strongly to DNA-ligase IV (21). XRCC7 encodes a These oligonucleotides were hybridized to the T7 460 kDa catalytic subunit (DNA-PKcs) of DNA-PK (20, 21). promoter primer supplied in the manufacturer’s kit and XRCC5 and XRCC6 encode 86 and 70 kDa subunits, used as directed. respectively, of Ku autoantigen, a DNA end-binding protein Chemically Synthesized siRNA. We have also made and regulatory subunit of DNA-PK (20, 21). The protein siRNAs by chemical synthesis at Ambion’s siRNA oligo- kinase activity of DNA-PK is believed to be stimulated by its nucleotide synthesis facility. We used the following sense association with DNA-end-bound Ku (20, 21). Ku hetero- and antisense oligonucleotides for target sequences Ku70-3, dimer binds to the ends of DNA in a non–sequence- Ku70-4, and Ku70-5 to produce chemically synthesized dependent manner (20, 21). The DNA-binding activity of Ku siRNA against Ku70. requires reduced sulfhydryl groups in cell-free systems (22). Ku heterodimer contains 14 cysteine residues and 5 of these (a) the siRNA for Ku70-3 consists of hybridized sense are located in Ku70, which is in contact with the DNA 5V-UUCAGGUGACUCCUCCAGGTT-3Vand antisense binding site (22). In a recent report, we have shown that Ku 5V-CCUGGAGGAGUCACCUGAATT-3Voligonucleotides. protein function can be greatly decreased in glucose-6- (b) the siRNA for Ku70-4 consists of hybridized sense phosphate dehydrogenase–deficient Chinese hamster ova- 5V-UUCUCUUGGUAACUUUCCCTT-3V and antisense ry cells by Ku protein thiol oxidation (23). However, this 5V-GGGAAAGUUACCAAGAGAATT-3Voligonucleotides. biochemical approach is effective only in glucose-6-phos- (c) the siRNA for Ku70-5 consists of hybridized sense phate dehydrogenase–deficient cells. 5V-GAUGCCCUUUACUGAAAAATT-3V and antisense In this study, we used a novel molecular approach to 5V-UUUUUCAGUAAAGGGCAUCTT-3Voligonucleotides. specifically target Ku70 mRNA. We show that this approach inhibits the expression of Ku70 protein and RNase III Generated. We have also generated siRNAs causes an increased response of cancer cells to g radiation by enzymatic digestion of long in vitro –transcribed double- and topoisomerase II inhibitor. stranded RNA molecule by the bacterial enzyme, RNase III. The enzymatic digestion produces double-stranded siR- NAs corresponding to several different target sequences. A Materials and Methods mixture of several different siRNAs targeted to different Cells and Growth Medium target sequences in a gene is expected to be effective in HeLa and HCT116 cancer cells were obtained from the most cells. To produce the Ku70 long double-stranded f American Type Culture Collection (Manassas, VA). These RNA, we first amplified an 200 bp region of Ku70 using cells were grown in McCoy 5A medium with 10% FCS and the primers 5V-AATTTAATACGACTCACTATAGGC- j V V 20 mmol/L HEPES at 37 C in a humidified 5% CO2 TATGGTACCGAGAAAGACA-3 and 5-AATTTAATAC- incubator. GACTCACTATAGGCTAAAGAGGTTGGCACAGAC-3V Ku70 Small Interfering RNA Synthesis with complementary DNA derived from HeLa cells. The In vitro –Transcribed siRNA. The nucleotide sequence 200 bp PCR product was run on a 1% agarose gel and of human Ku gene (accession #BC010034) was retrieved purified. The long double-stranded RNA and the siRNA from the National Center for Biotechnology Information. were then produced according to the instructions in the Twenty-one-mer nucleotides with starting amino acid siRNA cocktail kit (Ambion). nucleotides for several target sequences were selected siRNA #1 from Ambion (catalog #4850) was used in these using Ambion software. To produce in vitro –transcribed studies as negative control. small interfering RNA (siRNA) against Ku70, we used Ku70 siRNA Transfection the pSilencer siRNA construction kit according to the Forty thousand HeLa or HCT116 cells in a total volume of manufacturer’s recommendations (Ambion, Austin, TX). 370 AL McCoy 5A medium were plated in 24-well plates and The sense and antisense for two target sequences Ku70-3 incubated for 24 hours. The McCoy 5A medium was and Ku70-4 for in vitro –transcribed siRNA are shown aspirated and cells were rinsed with Earle’s balanced salt below: solution with 20 mmol/L HEPES. The cells were replen- ished with 200 AL of fresh DMEM and incubated for 1 hour Ku70-3 j at 37 C in a humidified 5% CO2 incubator. Cells were Sense 5-AATTCAGGTGACTCCTCCAGGV treated with Ku70 siRNA or scrambled siRNA as negative CCTGTCTC-3V control according to the manufacturer’s instructions. Briefly, Antisense 5-AACCTGGAGGAGTCACCTGAAV siRNA and OligofectAMINE were mixed separately with CCTGTCTC-3V OptiMem and incubated for 15 minutes at room tempera- and ture. These reagents were combined and incubated for another 15 minutes before adding to the cells in DMEM Ku70-4 without penicillin and streptomycin. The effect of Ku70 Sense 5V-AAGGGAAAGTTACCAAGAGAA siRNA on Ku70 protein was determined by Western blot CCTGTCTC-3V at 24, 48, and 72 hours after transfection. The effects of Antisense 5V-AATTCTCTTGGTAACTTTCCC g radiation and topoisomerase II poisons on HeLa cells were CCTGTCTC-3V estimated 72 hours after Ku70 siRNA transfection because

Mol Cancer Ther 2005;4(4). April 2005

72 hours’ incubation with siRNA showed the maximum 1 hour with secondary antibody (FITC-conjugated Affinity- inhibition of Ku70 expression. The effects of g radiation and Pure donkey anti-mouse IgG, catalog #715-095-150, Jackson topoisomerase II poisons on HCT116 cells were estimated ImmunoResearch, West Grove, PA) followed by washing 24, 48, and 72 hours after Ku70 siRNA transfection to with PBS and mounted using VectaShield with 4V,6- determine whether a direct correlation exists between Ku70 diamidino-2-phenylindole (Vector Labs, H1200). The cells expression and sensitization to radiation and etoposide. were visualized using an Olympus BX60 microscope and Western Blot Analysis quantified using Metamorph. The total amount of cellular Ku70 protein was quantified Radiation or EtoposideTreatment using Western blot and NIH image analysis. HeLa and The cells in 24-well plates were irradiated at room HCT116 cancer cells treated with and without Ku70 siRNA temperature in air using a Shepherd Mark I 137Cs irradiator. were rinsed with PBS thrice and mixed with 200 AL of lysis Radiation doses were as indicated in the figures. Alterna- buffer [20 mmol/L Tris (pH 7.6), 150 mmol/L NaCl, tively, they were incubated with different concentrations of 2 mmol/L EDTA, 10% glycerol, 1% Triton X-100, 1 mmol/L etoposide (Sigma) for 1 hour at 37jC in a humidified 5% DTT, 1 mmol/L orthovanadate, 2 mmol/L phenylmethyl- CO2 incubator. sulfonyl fluoride] containing 6 AL of protease inhibitor Clonogenic Assay cocktail (Sigma, St. Louis, MO) and 6 AL of NuPAGE The cells in 24-well plates were rinsed twice with 0.3 mL reducing agent. The cells in lysis buffer in the dish were of PBS and trypsinized with 0.1 mL of trypsin. The removed using a Teflon scraper and transferred to an trypsinized cells were resuspended in 0.2 to 0.3 mL of Eppendorf tube using a micropipette. The cells were medium and 0.1 mL was transferred to 12 mL of medium homogenized using a 1 mL syringe (22-gauge needle), and isotonic solution for plating and cell counting (Coulter centrifuged at 6,000 Â g for 2 minutes in a microfuge Counter), respectively. The cells were plated in 100 mm (Fisher 59A), and the supernatant stored at À80jC. The dishes at the required concentration to get no more than protein extracts (10 Ag) were incubated in NuPAGE sample 250 colonies per dish. The cells were cultured for 8 to 10 j buffer containing 4 AL of NuPAGE reducing agent at 70 C days in 5% CO2 to get viable colonies. A viable colony was for 10 minutes. These samples were then electrophoresed defined as having at least 50 cells after 10 days of growth. on a 10% Bis/Tris precast gel from NuPAGE at 200 V for The surviving fraction of treated cells was normalized to 60 minutes in MOPS SDS running buffer (NuPAGE). The the plating efficiency of control (untreated) cells. proteins were transferred to nitrocellulose by electropho- resis in NuPAGE transfer buffer at 20 V for 60 minutes. The nitrocellulose blot was incubated with 10 mL blocking Results buffer for 1.5 hours at room temperature on a rocker and To determine the effects of various siRNAs on Ku protein stored in the cold room overnight. The nitrocellulose paper expression, we measured the amount of Ku70 by Western was washed five times with PBS containing 0.1% Tween 20 blot analysis using a monoclonal anti-Ku70 antibody. A (PBST) and then incubated with 20 AL of primary Ku (p70) very slight change in the single f70 kDa immunoreac- antibody (Ab-4, clone N3H10, Neomarkers, Fremont, CA) tive band was observed in total cell lysates extracted in 10 mL of blocking buffer for 2 hours at room from HeLa cell lines treated for 24 hours with target temperature. The nitrocellulose paper was washed five sequences 4 (lanes 4 and 6) and 5 (lane 7) and RNase III times with PBST before incubation with secondary generated (lane 8) Ku70 siRNAs (Fig. 1). The Ku protein peroxidase-labeled mouse antibody for 1 hour, as per level was further decreased after 48 and 72 hours’ the manufacturer’s instruction (Amersham, Piscataway, incubation with siRNA with the highest decrease observed NJ), and the bands were detected using a standard after 72 hours. In addition, in vitro –transcribed siRNA for enhanced chemiluminescence kit. The density of bands in target sequence 3 (lane 3) also showed a significant decrease the resulting film was quantified using NIH image when cells were treated with siRNA for 48 and 72 hours. analysis. This indicates that 48 to 72 hours are required for the Immunofluorescence and Imaging maximum inhibition of Ku70 protein. The scrambled Immunofluorescence analysis was done as follows: first, siRNA treated HeLa cells (lane 2) had no measurable effect at the indicated time following transfection in a 24-well on the levels of Ku70. This siRNA was used as a negative tissue culture dish, the cells were fixed by 5-minute control for experiments that determined the sensitizing incubation in fresh 4% paraformaldehyde/PBS at room effects of Ku70 siRNA. Of the five Ku70 siRNAs, target temperature. The cells were washed with 1 mL of 1Â PBS, sequences 4 (lanes 4 and 6) and 5 (lane 7) are the most permeabilized by the addition of 0.1% Triton X-100/PBS effective in suppressing the expression of Ku70. The RNase and incubated for 5 minutes at room temperature. The cells III generated Ku70 siRNA (lane 8) inhibited the Ku70 were then washed with 1Â PBS, and blocked in 3% BSA/ expression less effectively than the Ku70 siRNA generated PBS for 1 hour and washed with 1Â PBS. Primary Ku70 for target sequences 3, 4, and 5. antibody (Ambion, catalog #4320) was diluted 1:200 in PBS, We have also determined the expression of Ku70 protein mixed with the cells, and incubated for 1 hour at room by immunohistochemistry in cells transfected with chem- temperature. The primary antibody was then removed and ically synthesized Ku70 siRNA for target sequence 5 the cells were washed with 1 mL of PBS, and incubated for because Western blot analysis indicated that this siRNA is

MolCancerTher2005;4(4).April2005

biphasic curve with a shoulder at low doses. This dose response curve is typical of mammalian cells. The HeLa cells treated with Ku70 siRNA for 72 hours were more sensitive to g radiation–induced cell killing than the control HeLa cells. To rule out the possibility that the radiosensitization effect of Ku70 siRNA may be nonspecific, we have also determined the effect of scrambled siRNA on the response of HeLa cells to g radiation. The results showed that the radiation response of HeLa cells transfected with scrambled siRNA was similar to that of the control.

Figure 1. Effects of various siRNAs on Ku70 expression as measured by Western blot analysis using a monoclonal anti-Ku70 antibody in HeLa cells. A, Western blot of cellular protein extracts of HeLa cells at 24, 48, and 72 h after transfection with Ku70 siRNA. Control untreated (lane 1), negative control siRNA (lane 2), in vitro – transcribed Ku70 siRNA for target sequence #3 (lane 3), in vitro – transcribed Ku70 siRNA for target sequence #4 (lane 4), chemically synthesized Ku70 siRNA for target sequence #3 (lane 5), chemically synthesized Ku70 siRNA for target sequence #4 (lane 6), chemically synthesized Ku70 siRNA for target sequence #5 (lane 7), and RNase III generated Ku70 siRNA (lane 8). B, graph of the effects of siRNA on total Ku protein quantitated by NIH image analysis plot of each lane in the gel and presented as percentage of untreated control. Each experiment was repeated at least thrice with SD as shown unless smaller than points plotted. The loading control actin indicated no significant difference in loading.

the most effective at 72 hours after transfection (Fig. 1B). We have normalized the Ku70 immunohistochemical staining with 4V,6-diamidino-2-phenylindole staining, and the amount of fluorescent signal from the nontransfected control (Fig. 2A). The relative fluorescent signal per cell was quantified using Metamorph and graphed (Fig. 2B). Ku70 siRNA for target sequence 5 has reduced the expression of Ku70 in HeLa cells to 20%, 40%, and 70% at 24, 48, and 72 hours after transfection. Of the five Ku70 siRNAs, we have tested the effect of chemically synthesized siRNAs for target sequence 5 on the g radiation response because it had shown the maximum Figure 2. Immunofluorescence image analysis of HeLa cells transfected inhibition of Ku70 expression at 72 hours after transfection with chemically synthesized Ku70 siRNA for target sequence 5. A, Ku70 (Fig. 1B). Figure 3 illustrates the effect of g radiation, in the expression in HeLa cells at 24, 48, and 72 h after transfection with Ku70 presence of air, on the clonogenic survival of HeLa cells 72 siRNA. The cells were visualized using an Olympus BX60 microscope. B, graph of the effects of Ku70 siRNA (relative fluorescence per cell hours after Ku70 siRNA transfection. The semi-log plot of compared with a nontransfected sample) on total Ku protein quantified dose-response curves for the control HeLa cells showed a by Metamorph analysis.

Mol Cancer Ther 2005;4(4). April 2005

72 hours were more sensitive to etoposide-induced cell killing than the control HeLa cells. The results with scrambled siRNA on the response of HeLa cells to etoposide showed that the etoposide response of HeLa cells transfected with scrambled siRNA was similar to that of the control. Treatment of the HCT116 cells with etoposide also resulted in a decrease in cell survival (Fig. 7A–C). However, the Ku70 siRNA-transfected HCT116 cells were much more sensitive to etoposide at 48 hours after transfection than the nontransfectant (Fig. 7B). Similar to the response of transfectants to radiation, HCT116 transfected with Ku70 siRNA for 24 and 72 hours did not show any significant sensitization or inhibition of Ku expression; suggesting a direct correlation between Ku70 inhibition and etoposide response (Fig. 7A–C).

Figure 3. Effects of g radiation, in the presence of air, on the clonogenic survival of HeLa cells 72 h after Ku70 siRNA transfection. Of the five Ku70 Discussion siRNAs synthesized, we have tested the effect of chemically synthesized Repair proteins have been shown to play a major role in siRNAs for target sequence 5 on the g radiation response because it had shown 70% inhibition of Ku70 expression at 72 h after transfection. Each the response of cells to DNA damaging agents (24, 25). experiment was repeated at least thrice with SD as shown unless smaller than points plotted.

To determine the generality of this Ku70 siRNA (CS #5), we have also tested the effects of this siRNA on colon carcinoma cells HCT116. The Western blot analysis, using the same monoclonal anti-Ku70 antibody used for HeLa cells, showed 16% inhibition in Ku70 in total cell lysates extracted from HCT116 treated for 24 hours (Fig. 4, lane 3). The Ku protein level was further decreased to 70% at 48 hours after transfection with Ku70 siRNA (Fig. 4, lane 6). However, the cells recovered completely at 72 hours (Fig. 4, lane 9) after transfection, suggesting that the maximum inhibitory effect of siRNA was observed at 48 hours after transfection (i.e., somewhat earlier than the HeLa cells). In order to determine the correlation between Ku70 expression and radiation response, we have exposed the HCT116 cells transfected with Ku70 siRNA (CS #5) to radiation (Fig. 5A–C). HCT116 transfected for 48 hours showed significant sensitization to radiation (Fig. 5B). However, cells transfected for 24 and 72 hours did not show any significant radiation sensitization (Fig. 5A and C). Thus, the radiation response paralleled the change in Ku70 expression. Another extensively studied DNA-damaging agent is etoposide. This drug causes DNA damage by inhibiting topoisomerase II resulting in DNA double-strand breaks (15, 24). Therefore, we measured the etoposide dose- Figure 4. Effects of siRNAs on Ku70 expression as measured by response curves for HeLa cell transfectants. We have again Western blot analysis using a monoclonal anti-Ku70 antibody in HCT116 selected to test the effect of chemically synthesized siRNAs cells. A, Western blot of cellular protein extracts of HCT116 cells at 24 for target sequence 5 (lane 7) on the etoposide response (lanes 1 – 3), 48 (lanes 4 – 6), and 72 h (lanes 7 – 9) after transfection with Ku70 siRNA. Control untreated (lanes 1, 4, and 7), negative control because it had shown the maximum inhibition of Ku70 siRNA (lanes 2, 5, and 8), chemically synthesized Ku70 siRNA for target expression at 72 hours after transfection (Fig. 1B). Figure 6 sequence #5 (lanes 3, 6, and 9). B, graph of the effects of siRNA on total shows the effect of 1-hour exposure of different concen- Ku protein quantitated by NIH image analysis plot of each lane in the gel trations of etoposide, in the presence of air, on the clonogenic and presented as a percentage of untreated control. Each experiment was repeated at least thrice with SD as shown unless smaller than points survival of HeLa cells 72 hours after Ku70 siRNA trans- plotted. The loading control actin indicated no significant difference in fection. The HeLa cells treated with Ku70 siRNA for loading.

MolCancerTher2005;4(4).April2005

Figure 5. Effects of g radiation, in the presence of air, on the clonogenic survival of colon carcinoma cells HCT116 at 24, 48, and 72 h after Ku70 siRNA transfection. We have tested the effect of chemically synthesized siRNAs for target sequence 5 on the g radiation response. Each experiment was repeated at least thrice with SD as shown unless smaller than points plotted.

However, because of the tremendous variety of damage in cellular response to topoisomerase II poison has been types and the complexity, and number of repair proteins shown using DNA double-strand break repair–deficient and their complexes, it has often been difficult to pinpoint XRS strains (15, 17, 32). However, a recent report showed the precise function and damage specificity of each protein. that the hypersensitivity is observed only in Ku and not This is particularly so because several of the repair mutants DNA-PKcs-deficient cells (33). Thus, the work presented were isolated functionally before the advent of modern here is consistent with the dominant effect of Ku70 molecular techniques. To help with this problem, various depletion inducing sensitivity to both ionizing radiation mutants have now been characterized. They often seem to and etoposide. lack the function of only one repair protein. It has been For mammalian cells, lack of Ku-86 was shown to cause shown that Ku heterodimer is involved in the repair of severe radiation sensitivity in the XRS5 hamster line and DNA double-strand breaks via nonhomologous end- this mutant has been characterized with many other DNA joining (26–30). Radiation produces free radicals by random radiolysis of water. Radiation is known to induce cell death primarily by the action of groups of hydroxyl radicals producing DNA double-strand breaks. Nonhomologous end-joining plays a major role in the repair of DNA double-strand breaks induced by g radiation (20, 21). Studies with mutant rodent cells with deficiency in DNA- PKcs, Ku86, and DNA ligase IV have shown that these proteins are essential for the nonhomologous end-joining repair pathway and survival after irradiation (20, 21). Of these three major DNA repair proteins, Ku plays a central role by sensing DNA lesions and coordinating various DNA repair proteins (15, 17, 18). Topoisomerase II poisons also induce DNA double- strand breaks leading to cell death similar to g radiation. Although radiation directly induces double-strand breaks, topoisomerase II poisons induce double-strand breaks by stabilizing the covalent intermediate (DNA/topoisomerase Figure 6. Effects of etoposide, in the presence of air, on the clonogenic II complex) of the topoisomerase II reaction (15, 17, 31). survival of HeLa cells 72 h after Ku70 siRNA transfection. Of the five Ku70 Nonhomologous end-joining is likely to be the major siRNAs synthesized, we have tested the effect of chemically synthesized pathway involved in the repair of DNA double-strand siRNAs for target sequence 5 on the etoposide response because it had shown 70% inhibition of Ku70 expression at 72 h after transfection. Each breaks produced by topoisomerase II poisons (15, 17, 32). experiment was repeated at least thrice with SD as shown unless smaller The role of nonhomologous end-joining repair protein Ku than points plotted.

Mol Cancer Ther 2005;4(4). April 2005

Figure 7. Effects of etoposide, in the presence of air, on the clonogenic survival of colon carcinoma cells HCT116 at 24, 48 and 72 h after Ku70 siRNA transfection. We have tested the effect of chemically synthesized siRNAs for target sequence 5 on etoposide response. Each experiment was repeated at least thrice with SD as shown unless smaller than points plotted.

damages, including sensitivity to topoisomerase II poisons working to make a stable siRNA. The efficiency of Ku70 (15, 17, 18, 32). Most investigators have focused on DNA depletion correlated exactly with changes in radiation and repair–deficient mutants to study the mechanisms of DNA etoposide sensitivity, both quantitatively and qualitatively. repair (16, 28–32) with only a few examples exploiting Our results suggest that Ku70 must be depleted by at least molecular targeting of repair proteins to increase the 70% to significantly change DNA damage response. response of cancer cells to DNA damaging agents (34). We have recently shown that g radiation-induced cell The recent discovery that small interfering RNAs are death could be enhanced by hydroxyethyldisulfide, a thiol effective in silencing gene expression in mammalian cells oxidant, in glucose-6-phosphate dehydrogenase–deficient has opened up new avenues to silence repair protein Chinese hamster ovary mutant cells (10). This modification expression (35, 36). This is important for many reasons. For in radiation response correlated well with defects in DNA example, even though Ku exists as a functional hetero- double-strand break repair and loss of Ku binding to DNA dimer, Ku70 mutants have not been identified. Further- ends (10, 14). Two-log higher cell killing and 70% inhibition more, there is enormous difference in the cellular of DNA double-strand break repair and Ku binding in concentration of repair proteins, with human cells typically glucose-6-phosphate dehydrogenase mutant Chinese ham- having more Ku and DNA-PK than hamster cells. It is not ster ovary cells suggested that inhibition of Ku protein known why this should be so, particularly because human function will be highly effective in sensitizing the cancer cells are commonly more sensitive than their hamster cells to g radiation and other DNA-damaging chemother- counterparts. Our results have now shown that Ku70 apeutic agents (10, 14). However, hydroxyethyldisulfide- siRNA can be used to inhibit Ku protein synthesis in mediated loss of Ku function resulting from protein thiol human cancer cells and that this causes sensitization to modification is effective only in glucose-6-phosphate ionizing radiation and etoposide. This result is perhaps dehydrogenase–deficient cells (10, 14). Our results have unexpected, because even at the maximum inhibition of shown, for the first time, that transfection of HeLa and Ku70 reported here (f70% depletion), the cells contain HCT116 cancer cells with Ku70 siRNA significantly much more Ku than normal hamster cells. enhanced the response to g radiation and the chemother- There was considerable variation in the efficiency of the apy agent etoposide. various siRNA’s made. Only three of the five Ku70 These current results suggest that targeting of Ku protein siRNAs were highly effective in silencing the Ku protein by Ku70 siRNA is a possible approach to cancer therapy. It expression. The RNase III generated Ku70 siRNAs also has also raised possibilities to target other DNA repair decreased the Ku protein but at a much lower efficiency. proteins either alone or in combination with Ku70 siRNA to It will obviously be of interest to continue the search for increase the response of cancer cells to DNA-damaging more efficient Ku70 depletion, and we are currently agents (36).

MolCancerTher2005;4(4).April2005

Acknowledgments 18. Kemp LM, Sedgwick SG, Jeggo PA. X-ray sensitive mutants of CHO cells defective in double strand break rejoining. Mutat Res 1984;132: We thank Tim Norton (University of Pennsylvania) and Angie Cheng 189 – 96. (Ambion, Inc.) for their technical help. 19. Taccioli GE, Gottlieb TM, Blunt T, et al. Ku80 product of the XRCC5 gene and its role in DNA repair and V(D)J recombination. Science 1994; References 265:1442 – 5. 1. Coleman CN. Molecular biology in radiation oncology. Radiation 20. Featherstone C, Jackson SP. Ku, a DNA repair protein with multiple oncology perspective of BRCA1 and BRCA2. Acta Oncol 1999;38:55 – 9. cellular functions. Mutat Res 1999;434:3 – 15. 2. Grunbaum U, Meye A, Bache M, et al. Transfection with mdm2- 21. Jeggo PA. Identification of genes involved in repair of DNA double- antisense or wtp53 results in radiosensitization and an increased strand breaks in mammalian cells. Radiat Res 1998;150:S80 – 91. apoptosis of a soft tissue sarcoma cell line. Anticancer Res 2001;21: 2065 – 71. 22. Zhang WW, Yaneva M. Reduced sulfhydryl groups are required for DNA binding of Ku protein. Biochem J 1993;293:769 – 74. 3. Harries M, Kaye SB. Recent advances in the treatment of epithelial ovarian cancer. Expert Opin Investig Drugs 2001;10:1715 – 24. 23. Ayene IS, Stamato TD, Mauldin SK, et al. Mutation in the glucose-6- phosphate dehydrogenase gene leads to inactivation of Ku DNA end 4. Murren JR. Rationale for non-platinum chemotherapy in advanced binding during oxidative stress. J Biol Chem 2002;277:9929 – 35. NSCLC. Oncology 2001;15:29 – 34. 5. Kaufmann SH. Cell death induced by topoisomerase-targeted drugs: 24. Pommier Y. DNA topoisomerases and their inhibition by anthracy- more questions than answers. Biochim Biophys Acta 1998;1400: clines. In: Priebe W, editor. Anthracycline antibiotics. Washington, DC: 195 – 211. Am Chem Soc; 1995. p. 183 – 203. 6. Newton HB. Chemotherapy for the treatment of metastatic brain 25. Jeggo PA, Kemp LM. X-ray sensitive mutants of Chinese hamster tumors. Expert Rev Anticancer Ther 2002;2:495 – 506. ovary cell lines: isolation and cross-sensitivity to other DNA-damaging agents. Mutat Res 1983;112:313 – 27. 7. Ruckdeschel JC. Future directions in non-small-cell lung cancer: a continuing perspective. Oncology 1998;12:90 – 6. 26. Blunt T, Finnie NJ, Taccioli GE, et al. Defective DNA-dependent protein kinase activity is linked to V(D)J recombination and DNA 8. Grant S, Roberts JD. The use of cyclin-dependent kinase inhibitors repair defects associated with the murine scid mutation. Cell 1995; alone or in combination with established cytotoxic drugs in cancer 80:813 – 23. chemotherapy. Drug Resist Updat 2003;6:15 – 26. 27. Smider V, Rathmell WK, Lieber MR, Chu G. Restoration of X-ray 9. Baker A, Payne CM, Briehl MM, Powis G. Thioredoxin, a gene found resistance and V(D)J recombination in mutant cells by Ku cDNA. Science overexpressed in human cancer, inhibits apoptosis in vitro and in vivo. 1994;266:288 – 91. Cancer Res 1997;57:5162 – 7. 28. Critchlow SE, Bowater RP, Jackson SP. Mammalian DNA double- 10. Ayene IS, Koch CJ, Tuttle SW, Stamato TD, Perez ML, Biaglow JE. strand break repair protein XRCC4 interacts with DNA ligase IV. Curr Biol Oxidation of cellular thiols by hydroxyethyldisulfide inhibits DNA double 1997;7:588 – 98. strand break rejoining in G6PD deficient mammalian cells. Int J Radiat Biol 2000;76:1523 – 31. 29. Gottlieb TM, Jackson SP. The DNA-dependent protein kinase: requirement for DNA ends and association with Ku antigen. Cell 1993; 11. Bristow RG, Benchimol S, Hill RP. The p53 gene as a modifier of intrinsic radiosensitivity: implications for radiotherapy. Radiother Oncol 72:131 – 42. 1996;40:197 – 223. 30. Grawunder U, Wilm M, Wu X, et al. Activity of DNA ligase IV stimulated by complex formation with XRCC4 protein in mammalian cells. 12. Endo K, Kuratate I, Watanabe M, et al. Wild-type p53 gene Nature 1997;388:492 – 5. transfection in human cultured sarcomas: effect of CDDP. Oncol Rep 2001;8:637 – 42. 31. Li Z, Otevreil T, Gao Y, et al. The XRCC4 gene encodes a novel protein 13. Harney JV, Seymour CB, Murphy DM, Mothersill C. Variation in involved in DNA double strand break repair and V(D)J recombination. Cell the expression of p53, c-myc, and bcl-2 oncoproteins in individual 1995;83:1079 – 89. patient cultures of normal urothelium exposed to cobalt 60 g-rays and 32. Mimori T, Hardin JA. Mechanism of interaction between Ku protein N-nitrosodiethanolamine. Cancer Epidemiol Biomarkers Prev 1995; and DNA. J Biol Chem 1986;261:10375 – 9. 4:617 – 25. 33. Caplen NJ, Parrish S, Imani F, Fire A, Morgan RA. Specific inhibition 14. Ayene IS, Bernhard EJ, Muschel RJ, McKenna WG, Krisch RE, of gene expression by small double-stranded RNAs in invertebrate and Koch CJ. DNA as an important target in radiation induced apoptosis of vertebrate systems. Proc Natl Acad Sci U S A 2001;98:9742 – 7. myc and myc plus ras oncogene transfected cells. Int J Radiat Biol 34. Elbashir SM, Harborth J, Lendeckel W, Yalcin A, Weber K, Tuschl T. 2000a;76:343 – 55. Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured 15. Caldecott K, Banks G, Jeggo P. DNA double-strand break repair mammalian cells. Nature 2001;411:494 – 8. pathways and cellular tolerance to inhibitors of topoisomerase II. Cancer 35. Jin S, Inoue S, Weaver DT. Differential etoposide sensitivity of cells Res 1990;50:5778 – 83. deficient in the Ku and DNA-PKcs components of the DNA-dependent 16. Getts RC, Stamato TD. Absence of a Ku-like DNA end binding activity protein kinase. Carcinogenesis 1998;19:965 – 71. in the xrs double-strand DNA repair-deficient mutant. J Biol Chem 36. Peng Y, Zhang Q, Nagasawa H, Okayasu R, Liber HL, Bedford JS. 1994;269:15981 – 4. Silencing expression of the catalytic subunit of DNA-dependent protein 17. Jeggo PA, Caldecott K, Pidsley S, Banks GR. Sensitivity of Chinese kinase by small interfering RNA sensitizes human cells for radiation- hamster ovary mutants defective in DNA double strand break repair to induced chromosome damage, cell killing, and mutation. Cancer Res topoisomerase II inhibitors. Cancer Res 1989;49:7057 – 63. 2002;62:6400 – 44.

Mol Cancer Ther 2005;4(4). April 2005

Downloaded from mct.aacrjournals.org on September 29, 2021. © 2005 American Association for Cancer Research. Ku protein targeting by Ku70 small interfering RNA enhances human cancer cell response to topoisomerase II inhibitor and γ radiation

Iraimoudi S. Ayene, Lance P. Ford and Cameron J. Koch

Mol Cancer Ther 2005;4:529-536.

Updated version Access the most recent version of this article at: http://mct.aacrjournals.org/content/4/4/529

Cited articles This article cites 33 articles, 11 of which you can access for free at: http://mct.aacrjournals.org/content/4/4/529.full#ref-list-1

Citing articles This article has been cited by 5 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/4/4/529.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/4/4/529. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.