LABORATORY INVESTIGATION
ATM inhibitor KU-55933 increases the TMZ responsiveness
of only inherently TMZ sensitive GBM cell...
kinase related kinases (PIKK) [3]. These kinases initiate
cell cycle arrest through effects on Chk1 and Chk2 and
facilitat...
to respective wells, and the number of neurospheres formed
after 14 days of incubation were counted [5].
Results
Character...
phosphorylation of S345-Chk1 and T68-Chk2 in both
U251 (Fig. 3b) and U87 cells (Fig. 3c). In contrast, TMZ-
induced phosph...
positive for foci: 24 ± 1.5 % vs. 13.7 ± 0.5 %, respec-
tively, p = 0.02; U87 % nuclei positive for foci:
52.9 ± 1 % vs. 3...
Discussion
TMZ is an integral component of chemotherapy for glio-
blastoma (GBM). Unrepaired TMZ induced O6
methyl
guanine...
selective sensitizing effect in the tumor lines inherently
sensitive to TMZ. The extent of KU55933 mediated TMZ
sensitizat...
ATM inhibitors may selectively radiosensitize tumors
specifically with deficient p53 status. Our preliminary
results also su...
both established and primary GBM models provide a
framework for evaluation of ATM inhibitors currently in
development. Whi...
of 9

Nadkarni-KU55933-JournofNeuroonc

Published on: Mar 3, 2016
Source: www.slideshare.net


Transcripts - Nadkarni-KU55933-JournofNeuroonc

  • 1. LABORATORY INVESTIGATION ATM inhibitor KU-55933 increases the TMZ responsiveness of only inherently TMZ sensitive GBM cells Aditi Nadkarni • Meena Shrivastav • Ann C. Mladek • Paul M. Schwingler • Patrick T. Grogan • Junjie Chen • Jann N. Sarkaria Received: 1 September 2011 / Accepted: 21 September 2012 / Published online: 3 October 2012 Ó Springer Science+Business Media New York 2012 Abstract Ataxia telangiectasia mutated (ATM) kinase is critical in sensing and repairing DNA double-stranded breaks (DSBs) such as those induced by temozolomide (TMZ). ATM deficiency increases TMZ sensitivity, which suggests that ATM inhibitors may be effective TMZ sen- sitizing agents. In this study, the TMZ sensitizing effects of 2 ATM specific inhibitors were studied in established and xenograft-derived glioblastoma (GBM) lines that are inherently sensitive to TMZ and derivative TMZ-resistant lines. In parental U251 and U87 glioma lines, the addition of KU-55933 to TMZ significantly increased cell killing compared to TMZ alone [U251 survival: 0.004 ± 0.0015 vs. 0.08 ± 0.01 (p 0.001), respectively, and U87 sur- vival: 0.02 ± 0.005 vs. 0.04 ± 0.002 (p 0.001), respectively] and also elevated the fraction of cells arrested in G2/M [U251 G2/M fraction: 61.8 ± 1.1 % vs. 35 ± 0.8 % (p 0.001), respectively, and U87 G2/M frac- tion 25 ± 0.2 % vs.18.6 ± 0.4 % (p 0.001), respectively]. In contrast, KU-55933 did not sensitize the resistant lines to TMZ, and neither TMZ alone or combined with KU-55933 induced a G2/M arrest. While KU-55933 did not enhance TMZ induced Chk1/Chk2 activation, it increased TMZ- induced residual c-H2AX foci in the parental cells but not in the TMZ resistant cells. Similar sensitization was observed with either KU-55933 or CP-466722 combined with TMZ in GBM12 xenograft line but not in GBM12TMZ, which is resistant to TMZ due to MGMT overexpression. These find- ings are consistent with a model where ATM inhibition suppresses the repair of TMZ-induced DSBs in inherently TMZ-sensitive tumor lines, which suggests an ATM inhibitor potentially could be deployed with an improvement in the therapeutic window when combined with TMZ. Keywords Temozolomide Á Glioblastoma Á DNA repair Á ATM inhibitor Introduction Integration of small molecule DNA repair inhibitors into GBM therapy has the potential to enhance the efficacy of temozolomide (TMZ) and improve the outcome of GBM treatment [1]. The key cytotoxic DNA lesion induced by TMZ is O6-methylguanine, which is removed specifically by O6-methylguanine methyltransferase (MGMT) [2]. Disruption of MGMT-mediated repair ultimately can lead to stalled replication forks that degenerate into DNA dou- ble strand breaks (DSBs). These DSBs trigger a damage response mediated by ATM and the ATM and Rad3-related kinase (ATR) protein kinases phosphatidylinositol 30 Aditi Nadkarni and Meena Shrivastav contributed equally to this work. Electronic supplementary material The online version of this article (doi:10.1007/s11060-012-0979-0) contains supplementary material, which is available to authorized users. A. Nadkarni Á M. Shrivastav Á A. C. Mladek Á J. N. Sarkaria (&) Mayo Clinic, Guggenheim 6-01B, 200 1st Street SW, Rochester, MN 55905, USA e-mail: sarkaria.jann@mayo.edu P. M. Schwingler Eastern Virginia Medical School, Norfolk, VA 23507, USA P. T. Grogan University of Kansas Medical School, Lawrence, KS 66106, USA J. Chen University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA 123 J Neurooncol (2012) 110:349–357 DOI 10.1007/s11060-012-0979-0
  • 2. kinase related kinases (PIKK) [3]. These kinases initiate cell cycle arrest through effects on Chk1 and Chk2 and facilitate the assembly and activation of DNA repair complexes to restore DNA integrity. Consistent with a critical role in DNA repair, ATM inactivation is associated with increased sensitivity to ionizing radiation and other DSB-inducing agents [4]. Following TMZ treatment, ATM modulates the repair of secondary DSBs, and ATM defi- ciency is associated with increased sensitivity to TMZ [3]. KU-55933 is a specific ATM inhibitor and a potent sen- sitizing agent when combined with radiation. The speci- ficity of this compound for ATM was established by counter-screening it against other members of the PIKK family which demonstrated a 100-fold differential in selectivity towards ATM kinase activity. KU-55933 spe- cifically inhibited ATM-mediated DNA repair events [4] and also sensitized patient xenograft derived stem-like neurospheres to TMZ. Given the potential role for ATM in modulating the repair of secondary DSBs induced by TMZ, we tested the hypothesis that ATM inhibitors would enhance the efficacy TMZ in inherently TMZ-sensitive glioma cell lines in which TMZ treatment will result in DNA double-strand breaks, and compared the combination treatment in paired TMZ-resistant cell lines. Materials & methods Cell culture & antibodies U251 and U87 malignant glioma cell lines were maintained in DMEM (Life Technologies, Inc.) supplemented with 10 % fetal bovine serum, 1 % penicillin and 1 % strepto- mycin. U251 and U87 cells were cultured and passaged over 8 weeks in the presence of escalating concentrations of TMZ (30–300 microM) to generate TMZ resistant lines, which are denoted as U251TMZ and U87TMZ, respectively. Short-term explant cultures from the primary GBM12 xenograft line and a derivative resistant GBM12TMZ line were grown in Neurobasal media (Invitrogen catalog# A1050901) [5]. Antibodies specific for phospho-Chk1 (catalog #2341), phospho-Chk2 (catalog #2661), total Chk1 (catalog #2345), total Chk2 (catalog #2662), c-H2AX (cat- alog #2577) were obtained from Cell Signaling, and phos- pho-ATM (catalog #ab81292) and ATM (catalog #10939) were obtained from Abcam. KU-55933 was synthesized by Ryss Laboratories Inc. and CP466722 (catalog #S2245) was purchased from Selleck Chemicals. Cy-Quant cell proliferation assay U251 and U87 malignant glioma cell lines were plated at a density of 1,000 and 500 cells per well, respectively, in 96-well plates, treated with 10, 30, 100 and 300 microM TMZ for 6 days and processed per manufacturer’s instructions (Invitrogen, CA). Clonogenic assay The effect of KU-55933 on TMZ sensitivity of U87 and U251 parental and TMZ resistant cells was assessed in a clonogenic survival assay as previously described [6, 7]. Cells were treated with 10, 30 or 300 microM TMZ alone or with a 1 h pretreatment with 10 microM KU-55933 and cultured for 2 weeks. Resultant colonies were stained with Coomassie Blue and quantitated. Western blotting Cells were treated with 30 microM TMZ with or without 10 microM KU-55933 and processed as previously described [8]. The effects of graded concentration of KU- 55933 or CP466722 on ATM phosphorylation were eval- uated similarly. Immunoblotting was carried out using nitrocellulose membranes and the indicated antibodies as previously described [6, 7]. Flow cytometric analysis Cells treated with TMZ (10 microM) alone or in combi- nation with KU-55933 (10 microM) were trypsinized, pelleted, fixed in 70 % ethanol/PBS, stored at -20 °C and analyzed by flow cytometry as described [6]. Immunocytochemistry Cells cultured on coverslips were treated with 30 microM TMZ with or without 10 microM KU-55933 for 24 or 72 h, fixed with 4 % paraformaldehyde, permeabilized with 0.3 % Triton-X100 and stained overnight with a c-H2AX specific primary antibody (Cell Signaling, #2557). A rho- damine conjugated secondary antibody (Jackson Immunol- abs) and DAPI were used for the detection of foci and nuclei, respectively. Coverslips were mounted on slides with mounting media (Dako) and analyzed using a Zeiss LSM 510 Confocal Laser Scanning Microscope at 60X magnifi- cation and the number of nuclei positive for foci was quantitated. Neurosphere formation assays Short-term explant cultures from GBM12 and GBM12TMZ xenografts seeded in triplicate at (250 cells per well in a 96-well plate) and cultured in stem cell media (Invitrogen A1050901). Four hours post-seeding, TMZ, with or without either KU-55933 or CP-466722 were added 350 J Neurooncol (2012) 110:349–357 123
  • 3. to respective wells, and the number of neurospheres formed after 14 days of incubation were counted [5]. Results Characterization of parental and TMZ resistant cell lines TMZ resistant sub-lines of U251 and U87 cells were generated by exposure to escalating doses of TMZ (30, 100 and 300 microM) over a period of two months. Surviving colonies were pooled to establish the TMZ-resistant U251TMZ and U87TMZ cell lines, respectively. In a Cy- Quant cell proliferation assay, these lines were markedly more resistant to the clinically relevant dose range of 30-100 microM TMZ compared to the parental lines (Fig. 1a, b). Compared to control, treatment with 30 microM TMZ was associated with a relative absorbance of 38.2 ± 2.1 % in U251 cells versus 97.4 ± 4.8 % in U251TMZ cells (p 0.001) and 51.4 ± 3.4 % in U87 versus 94.1 ± 0.9 % in U87TMZ (p 0.001). The TMZ- induced DNA damage response also was characterized in these lines by flow cytometry and western blotting. Treatment with 30 microM TMZ resulted in a marked increase in the fraction of cells arrested in G2/M, compared to untreated cells 72 h after treatment for parental U251 (90.8 ± 3.3 % vs. 7.6 ± 1.4 %, respectively, p = 0.001) and U87 cells (84.6 ± 13 % vs. 7.9 ± 1 %, respectively, p = 0.001). In contrast, the TMZ resistant cells did not accumulate in G2/M following treatment with TMZ (Fig. 1c, d). Consistent with checkpoint activation, treat- ment of parental U251 and U87 cells with 30 microM TMZ resulted in prolonged induction of T68-Chk2 phosphory- lation in the parental U251 and U87 cells at 24, 72 and 144 h time points (Fig. 1e, f). Similarly, increased phos- phorylation of S345-Chk1 following TMZ treatment was observed at all 3 time points in U251 and only at 24 h in U87 cells. In contrast, TMZ treatment in U251TMZ and U87TMZ lines was associated with a lack of Chk1 phos- phorylation and marginal changes in Chk2 phosphorylation compared to untreated controls. Thus, in comparison to the parental lines, development of TMZ resistance in the U87TMZ and U251TMZ lines is associated with a loss of TMZ-induced G2/M arrest and associated checkpoint activation. ATM inhibitor KU-55933 sensitizes only parental GBM cell lines to TMZ The effects of KU-55933 on cell survival were examined using a clonogenic assay. Treatment with 10 microM KU- 55933 significantly sensitized U251 cells to TMZ (Fig. 2a; survival after 30 microM TMZ 0.08 ± 0.01 without KU- 55933 versus 0.004 ± 0.001 with KU-55933, p 0.001). U87 cells also were sensitized by KU-55933 treatment, although the extent of sensitization was less profound (Fig. 2b; survival after 30 microM TMZ 0.04 ± 0.002 without KU-55933 versus 0.02 ± 0.005 with KU-55933. p 0.001). In contrast, the KU-55933 did not sensitize either TMZ resistant line to TMZ (U251TMZ survival: 0.84 ± 0.03 vs. 0.87 ± 0.01, respectively, p [ 0.1, and U87TMZ survival: 0.62 ± 0.03 vs 0.63 ± 0.09, respec- tively, p [ 0.1). These data suggest that KU-55933 selec- tively sensitizes parental but not TMZ-resistant GBM cells to TMZ. Consistent with the selective sensitizing effects of KU- 55933 in the parental cells, KU-55933 increased TMZ- induced G2/M accumulation of cells compared to TMZ treatment alone. Both TMZ and TMZ ? KU-55933 treat- ments resulted in a significant accumulation of U251 cells at G2/M 72 h following treatment, but by 144 h after treat- ment, combined treatment with KU-55933 and TMZ was associated with a persistent G2/M arrest (61.8 ± 1.1 % cells in G2/M) as compared to treatment with TMZ alone (35 ± 0.8 % cells in G2/M, p 0.001; Fig. 2c). In U87 cells, the increased G2/M accumulation associated with combined TMZ/KU-55933 treatment compared to TMZ alone was observed both at 72 h (27.5 vs. 21.4 respectively; p = 0.007) and 144 h (25.7 vs. 18.7 respectively; p 0.001) (Fig. 2d). In contrast, co-treatment of the resis- tant lines with KU-55933 and TMZ did not result in an increase in the fraction of cells arrested in G2/M, as com- pared to monotherapy (U251TMZ G2/M fraction: 20 ± 0.6 % vs. 19.7 ± 1.9 % (p = 0.58), respectively and U87TMZ G2/M fraction 14 ± 3.14 % vs. 9.8 ± 1.9 %, (p = 0.2), respectively). Thus, the effects of KU-55933 on TMZ-induced G2/M arrest are significantly greater in the inherently sensitive U251 and U87 cells as compared to the TMZ-resistant lines. In conjunction with the cell cycle analysis, the effects of KU-55933 on TMZ-induced phosphorylation of ATM, Chk1 and Chk2 were characterized. Phosphorylation of Ser1981 on ATM has previously been reported as a marker of ATM activation, and in parental U251 cells, TMZ treatment induced ATM phosphorylation by 24 h with robust activation by 72 h (Fig. 3a). Interestingly, co-treat- ment with KU55933 resulted in only minimal suppression of phosphorylation at either site 24 h after treatment and had no effect at 72 h after treatment despite robust sup- pression of radiation-induced ATM phosphorylation. In contrast, TMZ treatment in U251TMZ cells resulted in a delayed and blunted ATM phosphorylation that was not reproducibly affected by KU55933 co-treatment. Similar to the ATM activation pattern, treatment with TMZ with or without KU-55933 for 72 h resulted in robust J Neurooncol (2012) 110:349–357 351 123
  • 4. phosphorylation of S345-Chk1 and T68-Chk2 in both U251 (Fig. 3b) and U87 cells (Fig. 3c). In contrast, TMZ- induced phosphorylation of either Chk1 or Chk2 was markedly attenuated with either treatment in the corre- sponding U251TMZ and U87TMZ lines, and no repro- ducible differences were observed between therapy with TMZ alone vs. TMZ ? KU-55933. These data suggest an attenuated damage signal induced by TMZ with or without KU-55933 in the TMZ-resistant cell lines. Influence of KU-55933 on DNA damage processing Unrepaired DSBs appear as residual punctuate nuclear c- H2AX foci indicative of incomplete or defective DSB repair [9]. Hence, c-H2AX was used to assess the repair of TMZ-induced DNA damage in the sensitive and resistant lines. KU-55933 increased TMZ-induced percent nuclei positive for c-H2AX foci 72 h after treatment compared to TMZ alone in the parental lines (Fig. 4-U251 % nuclei 200 300 U251 U251TMZ p=0.007* 144 24 72 144 U251TMZ - + - + - + - + U251TMZ U251 10 300 p= 0.39 0 100 0 30 60 90 120 TMZ (uM) %absorbance A E 24 72 U251 P-T68-CHK2 CHK1 CHK2 - + - +TMZ, 30 µM Time, hrs. P-S345-CHK1 0 25 50 75 100 TMZ, microM 10 300 * p= 0.001 %cellsinG2phase C 0 100 200 300 TMZ (uM) 0 30 60 90 120 U87 U87TMZ p<0.001* %absorbance B F 24 72 14424 72 144 P-T68-CHK2 CHK1 CHK2 TMZ, 30 µM Time, hrs. P-S345-CHK1 U87 U87TMZ - + - + - + - + - + - + 0 25 50 75 100 U87 U87TMZ * p= 0.02 TMZ, microM 10 300 10 300 p= 0.48 %cellsinG2phase D Fig. 1 Characterization of TMZ resistant GBM cell lines. a–b The indicated cell lines were plated in triplicates and exposed to graded concentrations of TMZ. After incubation for 144 h, cell survival was quantified using the CyQuant colorimetric assay. The mean relative absorbance ± SE from two independent experiments is presented, c– d Cells treated with or without 30 microM TMZ were fixed at 72 h and analyzed for cell cycle distribution by flow cytometric analysis of cellular DNA content. The percent of cells in G2/M (mean ± SEM from three independent experiments is presented, e–f Cells treated with or without TMZ were harvested at indicated time points, and phosphorylation and total levels of Chk1 and Chk2 were analyzed by western blot. Representative results from 3 independent experiments are shown 352 J Neurooncol (2012) 110:349–357 123
  • 5. positive for foci: 24 ± 1.5 % vs. 13.7 ± 0.5 %, respec- tively, p = 0.02; U87 % nuclei positive for foci: 52.9 ± 1 % vs. 32.5 ± 2.5 %, respectively, p = 0.01). In contrast, neither treatment with TMZ alone nor TMZ in combination with KU-55933 induced a significant increase in c-H2AX foci in U251TMZ and U87TMZ. Collectively, these results demonstrate that KU-55933 significantly enhances the number of residual DNA DSBs induced by TMZ only in the inherently sensitive U251 and U87 cell lines. Effect of ATM inhibitors on primary GBM xenograft lines The effects of ATM inhibition on TMZ sensitivity were extended by testing either KU-55977 or a second ATM inhibitor, CP-466722 [10], in short-term explant cultures derived from the primary GBM12 and derivative GBM12TMZ xenograft line. Treatment of GBM12 cells with either KU-55933 or CP-466722 sensitized GBM12 cells to TMZ (Fig. 5a; survival after 6 microM TMZ alone 0.49 ± 0.09 versus 0.32 ± 0.05 with TMZ and KU-55933, p = 0.06 or 0.20 ± 0.06 with TMZ and CP-466722, p = 0.02). In contrast, combination of TMZ with either KU-55933 or CP-466722 had no effect on neurosphere formation in the TMZ-resistant GBM12TMZ line (Fig. 5b; survival after 30 microM TMZ alone 1.02 ± versus 0.98 ± 0.08 with KU-55933, p = 0.5 or 0.96 ± 0.04 with KU-55933, p = 0.4). To test whether the lack of sensiti- zation in the GBM12TMZ resistant line was due to lack of ATM inhibition, the effects of either ATM inhibitor on radiation-induced ATM phosphorylation was assessed in GBM12TMZ neurospheres. As seen in Fig. 5c, both inhibitors effectively suppressed radiation induced Ser- 1981-ATM phosphorylation, suggesting that the lack of TMZ sensitization by ATM inhibitors in the TMZ-resistant neurospheres is specific to TMZ induced damage. Collec- tively, these data further support the idea that ATM inhibitors may enhance the efficacy of TMZ in tumors that are inherently sensitive to TMZ. 0 10 20 30 0.001 0.01 0.1 1 U251 U251 + KU55933 U251TMZ U251TMZ + KU55933 p<0.001* p= 0.28 TMZ (µM) Cellsurvival 0 10 20 30 0.01 0.1 1 U87 U87 + KU55933 U87TMZ U87TMZ + KU55933 p<0.001* p= 0.98 TMZ (µM) Cellsurvival A B C D 0 10 20 30 40 50 60 70 72 hours 144 hours U251TMZ U251 * TMZ, 30 µM KU-55933, 10 µM - - - + + + + - - - - + + + + - - - - + + + + - - - - + + + + - p= 0.00002 p= 0.045 * p= 0.8 p= 0.6 %cellsinG2phase TMZ, 30 µM KU-55933, 10 µM 0 10 20 30 * - - - + + + + - - - - + + + + - - - - + + + + - - - - + + + + - 72 hours 144 hours U87TMZ U87 * p= 0.0008p= 0.0067 p= 0.2 p= 0.1 %cellsinG2phase Fig. 2 KU55933 sensitizes TMZ responsive GBM cells to TMZ treatment. The effect of KU55933 on TMZ sensitivity of U251 (a) and U87 (b) cells was assessed in a clonogenic assay. Cell survival (mean ± SEM from three independent experiments) is plotted relative to TMZ dose for treatment with or without 10 microM KU55933. c–d Cells were collected at indicated time points after treatment with 30 microM TMZ alone or in combination with 10 microM KU-55933, fixed, and analyzed for cell cycle distribution by flow cytometric analysis of cellular DNA content. The percent of cells in G2/M (mean ± SEM from three independent experiments) is presented J Neurooncol (2012) 110:349–357 353 123
  • 6. Discussion TMZ is an integral component of chemotherapy for glio- blastoma (GBM). Unrepaired TMZ induced O6 methyl guanine (O6MG) lesions result in replication-associated DSBs which trigger cell death if left unrepaired [11]. Hence, targeting DSB repair pathways in combination with TMZ therapy may be a useful therapeutic strategy in GBM. The related kinases ATM, ATR, and DNA-PKcs play key roles in DNA repair and development of inhibitors for each of these kinases has been actively pursued. The ATM inhibitor, KU-55933, inhibits ATM at 100-fold lower concentrations compared to ATR or DNA-PKcs, and inhibits ATM-specific DNA repair events [4]. Similarly, CP-466722, an ATM-specific inhibitor identified by screening a targeted compound library, suppresses ATM- dependent phosphorylation and disrupts ATM function but does not affect ATR and DNA-PKcs activity [10]. In this study, both CP-466722 and KU-55933 enhanced the effi- cacy of TMZ only in inherently sensitive tumor lines and this effect was associated with increased DSB accumula- tion in inherently TMZ sensitive GBM tumor lines consistent with disruption of ATM-mediated DNA repair processes. These data support the concept of combining next generation ATM inhibitors with TMZ in a selected subset of tumors based on predicted TMZ sensitivity. The selective sensitization of inherently TMZ-sensitive tumor cells suggests that induction of DNA damage by TMZ may be important for the sensitizing effects of KU- 55933. TMZ treatment in parental tumor lines induced robust checkpoint activity up to 144 h after treatment and was associated with an increase in the formation of c- H2AX foci, which is indicative of DNA DSB induction. These data are consistent with a model in which O6MG lesions mis-pair with thymidine during replication and are recognized by the mismatch repair (MMR) pathway [3]. Resulting futile cycles of MMR result in stalled replication forks that subsequently can degenerate into lethal DSBs [3]. In contrast to the parental lines, treatment of the TMZ resistant lines was associated with a distinct lack of checkpoint activation or c-H2AX foci induction. Given the importance of ATM in regulating DNA damage repair, we hypothesize that inhibition of ATM suppresses DNA repair of the replication-induced DSBs, which can account for the A B C P-S1981-ATM ATM Harvest Time, Hr : 24 72 72* 0.5 24 72 72* 0.5 U251 U251TMZ KU-55933, 10 µM : - + - + - + - + - + - + - + - + - + - + - + - + - + - + TMZ, 30 µM : - - + + - - + + - - + + - - IR, 4Gy : - - - - - - - - - - - - + + - - + + - - + + - - + + - - - - - - - - - - - - - - + + - + - + - + - + - - + + - - + + U251 U251TMZ - - - + - + - + + - - + + KU-55933, 10 µM P-T68-CHK2 CHK1 CHK2 TMZ, 30 µM P-S345-CHK1 - + - + - + - + - - + + - - + + U87 U87TMZ - + - + - + - + - + + - - + + KU-55933, 10 µM P-T68-CHK2 CHK1 CHK2 TMZ, 30 µM P-S345-CHK1 Fig. 3 KU-55933 inhibits ATM-mediated phosphorylation events. a Cells treated with indicated concentrations of TMZ or radiation in combination with or without KU-55933, were harvested and processed for western blot analysis of phospho- and total levels of ATM at 24 and 72 h (* indicates KU-55933 was added a second time to the media 48 h after initial treatment). b–c Cells treated with 30 microM TMZ alone or in combination with 10 microM KU-55933 were harvested at 72 h and phospho- and total levels of Chk1 and Chk2 were analyzed by western blot. Representative results from three independent experiments are presented where paired parental and TMZ resistant lines were run on the same gel but intermediate non-relevant lanes were cropped out of the image 354 J Neurooncol (2012) 110:349–357 123
  • 7. selective sensitizing effect in the tumor lines inherently sensitive to TMZ. The extent of KU55933 mediated TMZ sensitization was much lower in the parental U87 cells as compared to U251 cells. U87 is wild-type for p53 while U251 has mutant p53, and previous studies with caffeine at concentrations that inhibit ATM and ATR suggest that A B C F 0 10 20 30 U251 U251TMZ * - - - + + + + - - - - + + + + - TMZ, 30 µM KU-55933, 10 µM p=0.02 p=0.7 %nucleipositiveforfoci TMZ, 30 µM KU-55933, 10 µM 0 20 40 60 U87 U87TMZ * - - - + + + + - - - - + + + + - p=0.01 p=0.2 %nucleipositiveforfoci Fig. 4 Influence of KU55933 on the processing of TMZ induced DNA damage. a Seventy-two hours after TMZ treatment, unrepaired DSBs were visualized by staining c-H2AX foci (red) formed in the nuclei (blue). b–c Percent nuclei positive for foci (mean ± S.E from two independent experiments) are presented J Neurooncol (2012) 110:349–357 355 123
  • 8. ATM inhibitors may selectively radiosensitize tumors specifically with deficient p53 status. Our preliminary results also suggest that p53 status may be important for TMZ-sensitizing effects, although this needs to be addressed in future studies using isogenic cell lines wild- type or deficient for p53. The mechanism of TMZ resistance does not appear to have a significant impact on the lack of sensitizing effects of ATM inhibitors in our models. High level MGMT expression is a well-established mechanism of TMZ resistance, and resistance in GBM12TMZ is mechanisti- cally linked to MGMT over-expression [5]. In contrast, MGMT is not overexpressed in either U87TMZ or U251TMZ lines (supplemental data Fig. 1). While previ- ous studies have established an association between TMZ resistance and mutations in the MSH6 and MSH2 mis- match repair genes, sequencing of these 2 genes did not reveal any mutations unique to the TMZ resistant lines and associated with the emergence of TMZ resistance (data not shown). While these lines could harbor mutations in in the other MMR genes such as PMS2 and MLH1, at this point the mechanism of TMZ resistance in U87TMZ or U251TMZ is clearly different from GBM12TMZ. Regardless of the mechanism of TMZ resistance, the results in this study demonstrate, in 3 different glioma models that the sensitizing effects of ATM-inhibition are limited to the parental lines, and at least in U87 and U251, the sensitizing effects of combined treatment were associ- ated with significantly increased H2AX foci. These data suggest that the lack of sensitizing effects of KU55933 or CP-466722 in the resistant lines may be related to the absence of TMZ induced DSBs in these resistant lines. Although the classic description of the ATR and ATM signaling pathways place ATR upstream from Chk1 and ATM upstream from Chk2, there is significant cross-talk between the 2 pathways. Specifically relevant to the current studies, ATR can mediate Chk2 phosphorylation in response to replicative stress [3]. Recent studies demon- strate that ATR can phosphorylate ATM at Ser1981 in addition to autophosphorylation by ATM [12].This may explain why KU-55933 did not block either ATM P-Ser1981 or Chk2 phosphorylation at 72 h. Similarly, both ATR and ATM, can phosphorylate H2AX in response to damage; in the present study c-H2AX foci were increased following treatment with KU-55933, which is keeping with a role for ATR in H2AX phosphorylation following replicative DSBs [13]. Although there is func- tional overlap between ATM and ATR in sensing DSBs, ATM plays a separate role in the processing of DSBs through direct interaction with key DNA repair proteins [14]. Hence, while ATM inhibition by KU-55933 did not inhibit checkpoint activation or H2AX phosphorylation, the KU-55933-associated increase in TMZ-induced resid- ual c-H2AX foci is consistent with a model where ATM inhibition suppresses repair of TMZ induced replicative- induced DNA DSBs. The current prognosis of GBM patients is extremely poor with a 15 month median survival with combined radiation and TMZ therapy, and novel therapeutic strate- gies are desperately needed. The studies presented here in 1 1.2 1.4 Relativeneurosphere# Relativeneurosphere# 0.8 1.0 1.2A B 0 2 0.4 0.6 0.8 TMZ TMZ+KU-55933 0.2 0.4 0.6 TMZ TMZ+KU-55933 0 . TMZ, µM TMZ+CP-466722 0.0 0 3 10 300 2 4 6 TMZ, µM TMZ+CP-466722 P-S1981-ATM CP466722, µ KU-55933, µ M 1 3 10 - - - M - - - - - - - 1 3 10 IR, 4 Gy - + + + + + + + GBM12TMZC ATM Fig. 5 ATM-inhibition sensitizes TMZ responsive patient xenograft derived stem- like neurospheres to TMZ. a– b The effect of two ATM inhibitors, KU-55933 and CP466722 on the TMZ sensitivity of parental GBM12 and TMZ resistant GBM12TMZ neurospheres was assessed by a neurosphere formation assay. Neurosphere survival (Mean ± SEM from three independent experiments) is plotted relative to TMZ concentration with or without either KU-55933 (10 micromolar) or CP-466722 (3 micromolar). c Western blot analysis of P-ATM following treatment with graded concentrations of KU-55933 or CP466722 followed by irradiation (IR) 356 J Neurooncol (2012) 110:349–357 123
  • 9. both established and primary GBM models provide a framework for evaluation of ATM inhibitors currently in development. While the currently available ATM inhibi- tors, including KU-55933 and CP-466722, lack appropriate pharmacokinetic properties to allow in vivo efficacy stud- ies, our data suggests inherently TMZ-sensitive tumors are more likely to respond to combination of an ATM inhibitor with TMZ compared to TMZ resistant or recurrent tumors. Multiple studies have demonstrated that tumor MGMT promoter hypermethylation is associated with increased responsiveness to TMZ, although these patients still have an ultimately very poor prognosis [11]. This suggests that MGMT hypermethylated tumors may benefit most from an ATM-inhibitor based TMZ sensitizing strategy. Normal tissues generally express high-level MGMT and are resis- tant to TMZ compared to TMZ-sensitive tumor cells that have suppressed MGMT expression [15]. Hence, an ATM inhibitor may effectively sensitize this subset of MGMT- methylated tumors to TMZ without marked sensitization of the bone marrow cells. This is in contrast to previous attempts to employ an MGMT inhibitor O6-benzyl guanine (O6BG) as a TMZ-sensitizing agent in which combinations with TMZ was associated with enhanced myelotoxicity [16]. In conclusion, selective TMZ sensitization by KU55933 or CP-466722 of inherently sensitive cells pro- vides a strong rationale for testing next generation ATM inhibitors in GBM models with differing sensitivities to TMZ. Acknowledgments The authors thank James Tarara and the Mayo Flow Cytometry and Optical Morphology Core personnel. Funding: Grant support provided by NIH SPORE grant CA108961, NIH RO1 CA127716, and the Brain Tumor Funder’s Consortium (all to JNS) Conflict of interest JNS is a recipient of a research grant from Merck Pharmaceuticals for unrelated research References 1. Helleday T et al (2008) DNA repair pathways as targets for cancer therapy. Nat Rev Cancer 8(3):193–204 2. Kaina B et al (2001) BER, MGMT, and MMR in defense against alkylation-induced genotoxicity and apoptosis. Prog Nucleic Acid Res Mol Biol 68:41–54 3. Caporali S et al (2004) DNA damage induced by temozolomide signals to both ATM and ATR: role of the mismatch repair system. Mol Pharmacol 66(3):478–491 4. Hickson I et al (2004) Identification and characterization of a novel and specific inhibitor of the ataxia-telangiectasia mutated kinase ATM. Cancer Res 64(24):9152–9159 5. Kitange GJ, Mladek AC, Carlson BL et al (2012) Inhibition of Histone deacetylation Potentiates the evolution of acquired Temozolomide resistance linked to MGMT upregulation in Glioblastoma Xenografts. Clin Cancer Res 6. Sarkaria JN et al (1998) Inhibition of phosphoinositide 3-kinase related kinases by the radiosensitizing agent wortmannin. Cancer Res 58(19):4375–4382 7. Eshleman JS et al (2002) Inhibition of the mammalian target of rapamycin sensitizes U87 xenografts to fractionated radiation therapy. Cancer Res 62(24):7291–7297 8. Alderton GK et al (2006) Regulation of mitotic entry by micro- cephalin and its overlap with ATR signalling. Nat Cell Biol 8(7):725–733 9. Banath JP, Klokov D, Macphail SH et al (2010) Residual gam- maH2AX foci as an indication of lethal DNA lesions. BMC Cancer 10:4 10. Rainey MD et al (2008) Transient inhibition of ATM kinase is sufficient to enhance cellular sensitivity to ionizing radiation. Cancer Res 68(18):7466–7474 11. Sarkaria JN et al (2008) Mechanisms of chemoresistance to alkylating agents in malignant glioma. Clin Cancer Res 14(10):2900–2908 12. Stiff T et al (2006) ATR-dependent phosphorylation and activa- tion of ATM in response to UV treatment or replication fork stalling. EMBO J 25(24):5775–5782 13. Ward IM, Chen J (2001) Histone H2AX is phosphorylated in an ATR-dependent manner in response to replicational stress. J Biol Chem 276(51):47759–47762 14. Cortez D et al (1999) Requirement of ATM-dependent phos- phorylation of brca1 in the DNA damage response to double- strand breaks. Science 286(5442):1162–1166 15. Hegi ME et al (2005) MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 352(10):997–1003 16. Quinn JA et al (2009) Phase II trial of temozolomide plus o6- benzylguanine in adults with recurrent, temozolomide-resistant malignant glioma. J Clin Oncol 27(8):1262–1267 J Neurooncol (2012) 110:349–357 357 123

Related Documents