Roles of DNA-dependent protein kinase and ATM in cell-cycle-dependent radiation sensitivity in human cells

The DNA Double-Strand Break Repair in Glioma: Molecular Players and Therapeutic Strategies

Gliomas are the most frequent type of tumor in the central nervous system, which exhibit properties that make their treatment difficult, such as cellular infiltration, heterogeneity, and the presence of stem-like cells responsible for tumor recurrence. The response of this type of tumor to chemoradiotherapy is poor, possibly due to a higher repair activity of the genetic material, among other causes. The DNA double-strand breaks are an important type of lesion to the genetic material, which have the potential to trigger processes of cell death or cause gene aberrations that could promote tumorigenesis. This review describes how the different cellular elements regulate the formation of DNA double-strand breaks and their repair in gliomas, discussing the therapeutic potential of the induction of this type of lesion and the suppression of its repair as a control mechanism of brain tumorigenesis.

Increased γ-H2A.X intensity in response to chronic medium-dose-rate γ-ray irradiation

Background

The molecular mechanisms of DNA repair following chronic medium-dose-rate (MDR) γ-ray-induced damage remain largely unknown.

Methodology/Principal Findings

We used a cell function imager to quantitatively measure the fluorescence intensity of γ-H2A.X foci in MDR (0.015 Gy/h and 0.06 Gy/h) or high-dose-rate (HDR) (54 Gy/h) γ-ray irradiated embryonic fibroblasts derived from DNA-dependent protein kinase mutated mice (scid/scid mouse embryonic fibroblasts (scid/scid MEFs)). The obtained results are as follows: (1) Automatic measurement of the intensity of radiation-induced γ-H2A.X foci by the cell function imager provides more accurate results compared to manual counting of γ-H2A.X foci. (2) In high-dose-rate (HDR) irradiation, γ-H2A.X foci with high fluorescence intensity were observed at 1 h after irradiation in both scid/scid and wild-type MEFs. These foci were gradually reduced through de-phosphorylation at 24 h or 72 h after irradiation. Furthermore, the fluorescence intensity at 24 h increased to a significantly greater extent in scid/scid MEFs than in wild-type MEFs in the G₁ phase, although no significant difference was observed in G₂/M-phase MEFs, suggesting that DNA-PKcs might be associated with non-homologous-end-joining-dependent DNA repair in the G₁ phase following HDR γ-ray irradiation. (3) The intensity of γ-H2A.X foci for continuous MDR (0.06 Gy/h and 0.015 Gy/h) irradiation increased significantly and in a dose-dependent fashion. Furthermore, unlike HDR-irradiated scid/scid MEFs, the intensity of γ-H2A.X foci in MDR-irradiated scid/scid MEFs showed no significant increase in the G₁ phase at 24 h, indicating that DNA repair systems using proteins other than DNA-PKcs might induce cell functioning that are subjected to MDR γ-ray irradiation.

Conclusions

Our results indicate that the mechanism of phosphorylation or de-phosphorylation of γ-H2A.X foci induced by chronic MDR γ-ray irradiation might be different from those induced by HDR γ-ray irradiation.

Chemosensitization of cancer cells by KU-0060648, a dual inhibitor of DNA-PK and PI-3K

DNA double-strand breaks (DSB) are the most cytotoxic lesions induced by topoisomerase II poisons. Nonhomologous end joining (NHEJ) is a major pathway for DSB repair and requires DNA-dependent protein kinase (DNA-PK) activity. DNA-PK catalytic subunit (DNA-PKcs) is structurally similar to PI-3K, which promotes cell survival and proliferation and is upregulated in many cancers. KU-0060648 is a dual inhibitor of DNA-PK and PI-3K in vitro. KU-0060648 was investigated in a panel of human breast and colon cancer cells. The compound inhibited cellular DNA-PK autophosphorylation with IC(50) values of 0.019 μmol/L (MCF7 cells) and 0.17 μmol/L (SW620 cells), and PI-3K-mediated AKT phosphorylation with IC(50) values of 0.039 μmol/L (MCF7 cells) and more than 10 μmol/L (SW620 cells). Five-day exposure to 1 μmol/L KU-0060648 inhibited cell proliferation by more than 95% in MCF7 cells but only by 55% in SW620 cells. In clonogenic survival assays, KU-0060648 increased the cytotoxicity of etoposide and doxorubicin across the panel of DNA-PKcs-proficient cells, but not in DNA-PKcs-deficient cells, thus confirming that enhanced cytotoxicity was due to DNA-PK inhibition. In mice bearing SW620 and MCF7 xenografts, concentrations of KU-0060648 that were sufficient for in vitro growth inhibition and chemosensitization were maintained within the tumor for at least 4 hours at nontoxic doses. KU-0060648 alone delayed the growth of MCF7 xenografts and increased etoposide-induced tumor growth delay in both in SW620 and MCF7 xenografts by up to 4.5-fold, without exacerbating etoposide toxicity to unacceptable levels. The proof-of-principle in vitro and in vivo chemosensitization with KU-0060648 justifies further evaluation of dual DNA-PK and PI-3K inhibitors.

Participation of DNA-PKcs in DSB repair after exposure to high- and low-LET radiation

Cellular lesions (e.g. DSBs) are induced into DNA upon exposure to radiation, with DSB complexity increasing with radiation ionization density. Using M059K and M059J human glioblastoma cells (proficient and deficient in DNA-PKcs activity, respectively), we investigated the repair of DNA damage, including DSBs, induced by high- and low-LET radiation [gamma rays, alpha particles and high-charge and energy (HZE) ions]. In the absence of DNA-PKcs activity, less DSB repair and increased recruitment of RAD51 was seen at 24 h. After exposure to (56)Fe heavy ions, the number of cells with RAD51 tracks was less than the number of cells with gamma-H2AX at 24 h with both cell lines. Using alpha particles, comparable numbers of cells with visible gamma-H2AX and RAD51 were seen at 24 h in both cell lines. M059J cells irradiated with alpha particles accumulated in S phase, with a greater number of cyclin A and RAD51 co-stained cells seen at 24 h compared with M059K cells, where an S-phase block is absent. It is proposed that DNA-PKcs plays a role in the repair of some frank DSBs, which are longer-lived in NHEJ-deficient cells, and some non-DSB clustered damage sites that are converted into DSBs at replication as the cell cycles through to S phase.

Low and high LET radiation‐induced apoptosis in M059J and M059K cells

PURPOSE

To investigate and compare the ability of DNA-dependent protein kinase (DNA-PK)-deficient and -proficient cells to undergo apoptosis after exposure to low and high linear energy transfer (LET) radiation.

MATERIALS AND METHODS

A human glioma cell line M059J lacking the catalytic subunit of DNA-PK (DNA-PKcs) and its DNA-PKcs-proficient counterpart, M059K, were exposed to 1 and 4 Gy of accelerated nitrogen ions (14N, 140 eV nm(-1), 8-12 Gy min(-1)) or 60Co gamma-rays (0.2 eV nm(-1), 0.7 Gy min(-1)). The induction of apoptosis was studied up to 144 h post-irradiation using two different methods: morphological characterization of apoptotic cells after fluorescent staining and cell size distribution analysis to detect apoptotic bodies. In parallel, protein expression of DNA-PKcs and poly(ADP-ribose) polymerase (PARP) as well as DNA-PK and caspase-3 activity were investigated.

RESULTS

Low and high LET radiations (4 Gy) induced a time-dependent apoptotic response in both cell lines. Low LET radiation induced a significantly elevated apoptotic response in M059J as compared with M059K cells at 144 h post-irradiation. Following high LET radiation exposure, there was no difference between the cell lines at this time. PARP cleavage was detected in M059J cells following both low and high LET irradiation, while only high LET radiation induced PARP cleavage in M059K cells. These cleavages occurred in the absence of caspase-3 activation.

CONCLUSIONS

M059J and M059K cells both display radiation-induced apoptosis, which occur independently of caspase-3 activation. The apoptotic course differs between the two cell lines and is dependent on the quality of radiation.

Role of DNA double-strand break repair genes in cell proliferation under low dose-rate irradiation conditions

Radiation-induced DNA double-stand breaks (DSBs) lead to numerous biological effects. To elucidate the molecular mechanisms involved in cellular responses to low dose and low dose-rate radiation, it is informative to clarify the roles of DSB repair related genes. In higher vertebrate cells, there are at least two major DSB repair pathways, namely non-homologous end-joining (NHEJ) and homologous recombination (HR). Here, it is shown that in chicken DT40 cells irradiated with gamma-rays at a low dose-rate (2.4 cGy/day), the growth delay in NHEJ-related KU70- and PRKDC (encoding DNA-PKcs)-defective cells were remarkably higher than in cells defective for the HR-related RAD51B and RAD54 genes. DNA-PKcs- defective human M059J cells also showed an obvious growth delay when compared to control M059K cells. RAD54(-/-)KU70(-/-) cells demonstrated their highest degree of growth delay after an X-irradiation with a high dose-rate of 0.9 Gy/min. However they showed a lower degree of growth delay than that seen in KU70(-/-) and PRKDC(-/-/-) cells exposed to low dose-rate irradiation. These findings indicate that cellular responses to low dose-rate radiation are remarkably different from those to high dose-rate radiation. The fact that both DT40 and mammalian NHEJ-defective cells were highly sensitive to low dose-rate radiation, provide a foundation for the concept that NHEJ-related factors may be useful as molecular markers to predict the sensitivity of humans to low dose-rate radiation.

Preclinical evaluation of a potent novel DNA-dependent protein kinase inhibitor NU7441

DNA double-strand breaks (DSB) are the most cytotoxic lesions induced by ionizing radiation and topoisomerase II poisons, such as etoposide and doxorubicin. A major pathway for the repair of DSB is nonhomologous end joining, which requires DNA-dependent protein kinase (DNA-PK) activity. We investigated the therapeutic use of a potent, specific DNA-PK inhibitor (NU7441) in models of human cancer. We measured chemosensitization by NU7441 of topoisomerase II poisons and radiosensitization in cells deficient and proficient in DNA-PK(CS) (V3 and V3-YAC) and p53 wild type (LoVo) and p53 mutant (SW620) human colon cancer cell lines by clonogenic survival assay. Effects of NU7441 on DSB repair and cell cycle arrest were measured by gammaH2AX foci and flow cytometry. Tissue distribution of NU7441 and potentiation of etoposide activity were determined in mice bearing SW620 tumors. NU7441 increased the cytotoxicity of ionizing radiation and etoposide in SW620, LoVo, and V3-YAC cells but not in V3 cells, confirming that potentiation was due to DNA-PK inhibition. NU7441 substantially retarded the repair of ionizing radiation-induced and etoposide-induced DSB. NU7441 appreciably increased G(2)-M accumulation induced by ionizing radiation, etoposide, and doxorubicin in both SW620 and LoVo cells. In mice bearing SW620 xenografts, NU7441 concentrations in the tumor necessary for chemopotentiation in vitro were maintained for at least 4 hours at nontoxic doses. NU7441 increased etoposide-induced tumor growth delay 2-fold without exacerbating etoposide toxicity to unacceptable levels. In conclusion, NU7441 shows sufficient proof of principle through in vitro and in vivo chemosensitization and radiosensitization to justify further development of DNA-PK inhibitors for clinical use.

A role for PKCzeta in potentiation of the topoisomerase II activity and etoposide cytotoxicity by wortmannin

Enhanced cytotoxicity of etoposide by wortmannin, an inhibitor of enzymes holding a phosphatidylinositol 3-kinase domain, was investigated in eight cell lines proficient or deficient for DNA double-strand break repair. Wortmannin stimulated the decatenating activity of topoisomerase II, promoted etoposide-induced accumulation of DNA double-strand breaks, shifted the specificity for cell killing by etoposide from the S to G1 phase of the cell cycle, and potentiated the cytotoxicity of etoposide through two mechanisms. (a) Sensitization to high, micromolar amounts of etoposide required integrity of the nonhomologous end-joining repair pathway. (b) Wortmannin dramatically increased the susceptibility to low, submicromolar amounts of etoposide in a large fraction of the cell population irrespective of the status of ATM, Ku86, and DNA-PKCS. It is shown that this process correlates depression of phosphatidylinositol 3-kinase-dependent phosphorylation of the atypical, zeta isoform of protein kinase C (PKCzeta). Stable expression of a dominant-negative, kinase-dead mutant of PKCzeta in a tumor cell line reproduced the hypersensitivity pattern induced by wortmannin. The results are consistent with up-regulation of the topoisomerase II activity in relation to inactivation of PKCzeta and indicate that PKCzeta may be a useful target to improve the efficiency of topoisomerase II poisons at low concentration.

Preclinical pharmacokinetics and metabolism of a novel prototype DNA-PK inhibitor NU7026

In this study we investigated the in vitro time dependence of radiosensitisation, pharmacokinetics and metabolism of NU7026, a novel inhibitor of the DNA repair enzyme DNA-dependent protein kinase (DNA-PK). At a dose of 10 μM, which is nontoxic to cells per se, a minimum NU7026 exposure of 4 h in combination with 3 Gy radiation is required for a significant radiosensitisation effect in CH1 human ovarian cancer cells. Following intravenous administration to mice at 5 mg kg⁻¹, NU7026 underwent rapid plasma clearance (0.108 l h⁻¹) and this was largely attributed to extensive metabolism. Bioavailability following interperitoneal (i.p.) and p.o. administration at 20 mg kg⁻¹ was 20 and 15%, respectively. Investigation of NU7026 metabolism profiles in plasma and urine indicated that the compound undergoes multiple hydroxylations. A glucuronide conjugate of a bis-hydroxylated metabolite represented the major excretion product in urine. Identification of the major oxidation site as C-2 of the morpholine ring was confirmed by the fact that the plasma clearance of NU7107 (an analogue of NU7026 methylated at C-2 and C-6 of the morpholine ring) was four-fold slower than that of NU7026. The pharmacokinetic simulations performed predict that NU7026 will have to be administered four times per day at 100 mg kg⁻¹ i.p. in order to obtain the drug exposure required for radiosensitisation.

Different G2/M accumulation in M059J and M059K cells after exposure to DNA double-strand break-inducing agents

PURPOSE

To investigate and compare the cell cycle progression in relation to cell death in the human glioma cell lines, M059J and M059K, after exposure to DNA double-strand break-inducing agents.

METHODS AND MATERIALS

The M059J and M059K cells, deficient and proficient in the catalytic subunit of the DNA-dependent protein kinase, respectively, were exposed to 1 and 4 Gy of photons or accelerated nitrogen ions. In addition, M059J and M059K cells were treated with 10 and 40 mug/mL of bleomycin for 30 min, respectively. Cell cycle progression, monitored by DNA flow cytometry, was measured up to 72 h after treatment.

RESULTS

M059J, but not M059K, cells displayed G(2)/M accumulation after low linear energy transfer irradiation. High linear energy transfer radiation exposure however, resulted in a substantial increase of M059K cells in the G(2)/M phase detected at 48 h. At 72 h, the number of cells in the G(2)/M phase was equivalent to its control. M059J cells accumulated mainly in S phase after high linear energy transfer irradiation. In contrast to M059K, M059J cells were still blocked at 72 h. Bleomycin induced G(2)/M accumulation for both M059J and M059K cells detected 24 h after treatment. At 48 h, the percentage of bleomycin-treated M059J cells in G(2)/M phase remained high, and the number of M059K cells had decreased to control levels. Neither cell line showed cell cycle arrest (< or =10 h) after exposure to these agents.

CONCLUSION

Distinct cell cycle block and release is dependent on the complexity of the induced DNA damage and the presence of the DNA-dependent protein kinase catalytic subunit.

Potentiality of DNA-dependent protein kinase to phosphorylate Ser46 of human p53

DNA damage induces accumulation and activation of p53 via various posttranslational modifications. Among them, several lines of evidence indicated the phosphorylation of Ser46 as an important mediator of DNA damage-induced apoptosis but the responsible kinase remains to be clarified, especially in the case of ionizing radiation (IR). Here we showed that DNA-dependent protein kinase (DNA-PK) could phosphorylate Ser46 of p53 in addition to reported phosphorylation sites Ser15 and Ser37. However, IR-induced phosphorylation of Ser46 was seen even in M059J, a human glioma cell line lacking DNA-PKcs, and it was, at most, only slightly less than in control M059K. On the other hand, a related kinase ataxia-telangiectasia mutated (ATM), which was shown to be essential for IR-induced phosphorylation of Ser46, could poorly phosphorylate Ser46 by itself. These results collectively suggested two pathways for IR-induced phosphorylation of Ser46, i.e., direct phosphorylation by DNA-PK and indirect phosphorylation via ATM.

Radiosensitization by hyperthermia in the chicken B-lymphocyte cell line DT40 and its derivatives lacking nonhomologous end joining and/or homologous recombination pathways of DNA double-strand break repair

Hyperthermia has a radiosensitizing effect, which is one of the most important biological bases for its use in cancer therapy with radiation. Although the mechanism of this effect has not been clarified in molecular terms, possible involvement of either one or both of two major DNA double-strand break (DSB) repair pathways, i.e. nonhomologous end joining (NHEJ) and homologous recombination (HR), has been speculated. To test this possibility, we examined cells of the chicken B-lymphocyte cell line DT40 and its derivatives lacking NHEJ and/or HR: KU70(-/-), DNA-PKcs(-/-/-), RAD54(-/-) and KU70(-/-)/RAD54(-/-). Radiosensitization by hyperthermia could be seen in all of the mutants, including KU70(-/-)/RAD54(-/-), which lacked both NHEJ and HR. Therefore, radiosensitization by hyperthermia cannot be explained simply by its inhibitory effects, if any, on NHEJ and/or HR alone. However, in NHEJ-defective KU70(-/-) and DNA-PKcs(-/-/-), consisting of two subpopulations with distinct radiosensitivity, the radiosensitive subpopulation, which is considered to be cells in G(1) and early S, was not sensitized. Substantial sensitization was seen only in the radioresistant subpopulation, which is considered to be cells in late S and G(2), capable of repairing DSBs through HR. This observation did not exclude possible involvement of NHEJ in G(1) and early S phase and also suggested inhibitory effects of hyperthermia on HR. Thus partial contribution of NHEJ and HR in radiosensitization by hyperthermia, especially that depending on the cell cycle stage, remains to be considered.

Double strand break repair by homologous recombination is regulated by cell cycle-independent signaling via ATM in human glioma cells

To investigate double strand break (DSB) repair and signaling in human glioma cells, we stably transfected human U87 (ATM(+), p53(+)) glioma cells with a plasmid having a single I-SceI site within an inactive green fluorescent protein (GFP) expression cassette, allowing for the detection of homologous recombination repair (HRR) by GFP expression. HRR and nonhomologous end joining (NHEJ) were also determined by PCR. DSB repair was first detected at 12 h postinfection with an adenovirus expressing I-SceI with repair reaching plateau levels between 24 and 48 h. Within this time frame, NHEJ predominated over HRR in the range of 3-50-fold. To assess the involvement of ATM in DSB repair, we first examined whether ATM was associated with the DSB. Chromatin immunoprecipitation showed that ATM was present at the site of the DSB as early as 18 h postinfection. In cells treated with caffeine, an inhibitor of ATM, HRR was reduced, whereas NHEJ was not. In support of this finding, GFP flow cytometry demonstrated that caffeine reduced HRR by 90% under conditions when ATM kinase activity was inhibited. Dominant-negative ATM expressed from adenovirus inhibited HRR by 45%, also having little to no effect on NHEJ. Furthermore, HRR was inhibited by caffeine in serum-starved cells arrested in G(0)/G(1), suggesting that ATM is also important for HRR outside of the S and G(2) cell cycle phases. Altogether, these results demonstrate that HRR contributes substantially to DSB repair in human glioma cells, and, importantly, ATM plays a critical role in regulating HRR but not NHEJ throughout the cell cycle.

DNA damage-induced cell-cycle phase regulation of p53 and p21waf1 in normal and ATM-defective cells

The ATM-dependent accumulation of p53 and induction of p21waf1 are key events for G1 cell-cycle checkpoint arrest following DNA damage. In ATM-null AT cells, even though the p53 and p21waf1 responses are kinetically delayed and quantitatively reduced, the G1 checkpoint is virtually disrupted, suggesting that these proteins arrive too late in G1 to enforce the arrest. As the precise mechanism remains unclear, we examined the response to DNA double-strand breaks generated by gamma-radiation (IR), to determine if ATM deficiency affects the cell-cycle phase regulation of these molecules. We find that, after irradiation, whereas normal LCL-N cells markedly increase their levels of p53 in all phases of the cell cycle, AT cells fail to show any p53 increase in the G1 phase. In addition, whereas in LCL-N p21waf1 is induced in G1 and G2-M, in AT cells this induction is partly seen in G2-M, but not in G1, indicating a different cell-cycle phase regulation of p53 and p21waf1 as a result of ATM deficiency. The levels and catalytic activity of the p53-targeting kinases ATR and DNA-PK in LCL-N and AT cells are very similar throughout the cell cycle, both before and after IR, thus excluding a phase-specific activity for these kinases. Collectively, our findings demonstrate that, in ATM-deficient cells, the p53-dependent p21waf1 response to DNA damage is not only quantitatively reduced, but also specifically suppressed in the G1 phase, thus providing a mechanistic explanation for the severe disruption of the G1 checkpoint in AT cells.

DNA-PK: the major target for wortmannin-mediated radiosensitization by the inhibition of DSB repair via NHEJ pathway

The effect of wortmannin posttreatment was studied in cells derived from different species (hamster, mouse, chicken, and human) with normal and defective DNA-dependent protein kinase (DNA-PK) activity, cells with and without the ataxia telangiectasia (ATM) gene, and cells lacking other regulatory proteins involved in the DNA double-strand break (DSB) repair pathways. Clonogenic assays were used to obtain all results. Wortmannin radiosensitization was observed in Chinese hamster cells (V79-B310H , CHO-K1), mouse mammary carcinoma cells (SR-1), transformed human fibroblast (N2KYSV), chicken B lymphocyte wild-type cells (DT40), and chicken Rad54 knockout cells (Rad54-/-). However, mouse mammary carcinoma cells (SX9) with defects in the DNA-PK and chicken DNA-PK catalytic subunit (DNA-PKcs) knockout cells (DNA-PKcs-/-/-) failed to exhibit wortmannin radiosensitization. On the other hand, SCID mouse cells (SC3VA2) exposed to wortmannin exhibited significant increases in radiosensitivity, possibly because of some residual function of DNA-PKcs. Moreover, the transformed human cells derived from AT patients (AT2KYSV) and chicken ATM knockout cells (ATM-/-) showed pronounced wortmannin radiosensitization. These studies demonstrate confirm that the mechanism underlying wortmannin radiosensitization is the inhibition of DNA-PK, but not of ATM, thereby resulting in the inhibition of DSB repair via nonhomologous endjoining (NHEJ).

Function of DNA-protein kinase catalytic subunit during the early meiotic prophase without Ku70 and Ku86

All components of the double-stranded DNA break (DSB) repair complex DNA-dependent protein kinase (DNA-PK), including Ku70, Ku86, and DNA-PK catalytic subunit (DNA-PKcs), were found in the radiosensitive spermatogonia. Although p53 induction was unaffected, spermatogonial apoptosis occurred faster in the irradiated DNA-PKcs-deficient scid testis. This finding suggests that spermatogonial DNA-PK functions in DNA damage repair rather than p53 induction. Despite the fact that early spermatocytes lack the Ku proteins, spontaneous apoptosis of these cells occurred in the scid testis. The majority of these apoptotic spermatocytes were found at stage IV of the cycle of the seminiferous epithelium where a meiotic checkpoint has been suggested to exist. Meiotic synapsis and recombination during the early meiotic prophase induce DSBs, which are apparently less accurately repaired in scid spermatocytes that then fail to pass the meiotic checkpoint. The role for DNA-PKcs during the meiotic prophase differs from that in mitotic cells; it is not influenced by ionizing radiation and is independent of the Ku heterodimer.