Suppression of the DNA damage response in acute myeloid leukemia versus myelodysplastic syndrome

Article metrics

Abstract

The molecular mechanisms responsible for the evolution from the preleukemic entities of low-risk myelodysplastic syndrome (MDS) to the less favorable forms of high-risk MDS, as well as those enabling transformation to acute myeloid leukemia (AML), are still incompletely understood. Abundant evidence from solid tumors demonstrates that preneoplastic lesions activate signaling pathways of a DNA damage response (DDR), which functions as an ‘anticancer barrier’ hindering tumorigenesis. Testing the hypothesis that subgroups of MDS and AML differ with respect to DDR, we first assessed markers of DDR (phosphorylation of ATM, Chk-1, Chk-2 and H2AX) in cell lines representing different entities of MDS (P39, MOLM-13) and AML (MV4-11, KG-1) before and after γ-irradiation. Although γ-irradiation induced apoptosis and G2/M arrest and a concomitant increase in the phosphorylation of ATM, Chk-1 and H2AX in MDS-derived cell lines, this radiation response was attenuated in the AML-derived cell lines. It is noteworthy that KG-1, but not P39 cells exhibit signs of an endogenous activation of the DDR. Similarly, we found that the frequency of P-ATM+ cells detectable in bone marrow (BM) biopsies increased in samples from patients with AML as compared with high-risk MDS samples and significantly correlated with the percentage of BM blasts. In contrast, the frequency of γ-H2AX+ cells was heterogeneous in all subgroups of AML and MDS. Whereas intermediate-1 MDS samples contained as little P-Chk-1 and P-Chk-2 as healthy controls, staining for both checkpoint kinases increased in intermediate-2 and high-risk MDS, yet declined to near-to-background levels in AML samples. Thus the activation of Chk-1 and Chk-2 behaves in accord with the paradigm established for solid tumors, whereas ATM is activated during and beyond transformation. In conclusion, we demonstrate the heterogeneity of the DDR response in MDS and AML and provide evidence for its selective suppression in AML because of the uncoupling between activated ATM and inactive checkpoint kinases.

Introduction

Myelodysplastic syndromes (MDS) are clonal hematopoietic stem cell disorders that are characterized by ineffective hematopoiesis leading to peripheral cytopenias (thrombopenia, leucopenia and anemia) and a frequent progression to acute myeloid leukemia (AML). MDS are classified based on the morphology and blast cell counts in the bone marrow (Bennett et al., 1982; Vardiman et al., 2002). The main prognostic factors of MDS, for progression to AML and survival, include the number of cytopenias, percentage of marrow blasts and cytogenetic abnormalities. These factors are combined in an International Prognostic Scoring System that distinguishes four subgroups with distinct probabilities of progression to AML and survival: low, intermediate 1 (int-1), intermediate 2 (int-2) and high. While the low and int-1 subgroups are often grouped together as ‘low-risk MDS’, the int-2 and high subgroups are condensed to ‘high-risk MDS’ (Greenberg et al., 1997).

Low-risk MDS are characterized by an increased apoptotic turnover of bone marrow progenitors, which accounts to a large extent for the cytopenias. In contrast, at later disease stages (high-risk MDS), a progressive increase in myeloblasts, coupled with a decrease in apoptosis, is seen and this progressive infiltration leads to cytopenias through marrow failure (Parker and Mufti, 1998, 2000, 2001). In these low-risk, as well as high-risk MDS patients, allogeneic stem cell transplantation remains the only option for curative treatment, yet is only applicable in 15–20% of all patients.

AML is highly heterogeneous in morphology, cytogenetics and prognosis. The single most important prognostic indicator in AML with respect to treatment efficacy, long-term remission and overall survival is the karyotype (Slovak et al., 2000; Grimwade, 2001; Grimwade et al., 2001, 2004). Accordingly, patients can be classified into three risk groups: (a) favorable, (b) intermediate, or (c) unfavorable. Whereas core-binding factor AML (that is AML with t(8;21) or inv(16)) and AML with t(15;17) constitute the favorable risk group, patients exhibiting a normal karyotype in the malignant clone are in the intermediate risk group. A complex karyotype, with three or more chromosome abnormalities, defines the unfavorable risk group (Klaus et al., 2004). As it is true in MDS, aggressive treatment regimens are associated with severe toxicity in AML and are thus frequently not applicable in an elderly patient population that often suffers from considerable co-morbidity.

The molecular mechanisms accounting for the transition from a pre-malignant condition with distinct degrees of risk of transformation (low-risk MDS, high-risk MDS) to full-blown neoplastic disease (AML) are poorly understood. There is abundant evidence for a variety of distinct solid tumors that oncogenesis follows a multi-step process of sequential (epi-)genetic alterations in which commonly a “barrier” against transformation has to be overcome. The activation of oncogenes commonly stimulates, perhaps as a result of replicative stress, a DNA damage response (DDR) that leads to a permanent cell cycle arrest (senescence) or apoptosis, thus avoiding the proliferation or survival of pre-malignant clones (Kastan and Bartek, 2004). This endogenous DDR involves the activation of DNA damage sensors (such as ATM), the activation of cell cycle checkpoint kinases (Chk-1, Chk-2) and the activation of tumor suppressor p53. Together with the phosphorylation of histone H2AX (which occurs within the so called γ-H2AX foci, also called ‘DNA damage foci’), these events can be monitored by means of phospho-neo-epitope-specific antibodies that detect the activating phosphorylation of ATM, Chk-1, Chk-2 or p53 (Shiloh, 2003; Gorgoulis et al., 2005; Bartkova et al., 2005a). As a rule, pre-malignant lesions (such as dysplastic epithelia) exhibit the presence of cells with multiple intranuclear γ-H2AX foci, as well as phosphorylated ATM, Chk-1, Chk-2 and/or p53. However, during or after the progression from in situ carcinomas to infiltrating tumors, the signs of the DDR vanish, presumably as a result of lacking oncogenic stress or due to inactivation (by mutation, silencing or inactivation of signaling cascades) of ATM, Chk-1, Chk-2 or p53 (DiTullio et al., 2002; Gorgoulis et al., 2005; Bartkova et al., 2005b). Hence, the protective ‘barrier’ against transformation has been lifted and the cancer can expand in an unrestrained fashion. Immunohistochemical studies on patient material revealed that this scenario applies to lung, breast, colon, urinary bladder and prostate carcinoma (Gorgoulis et al., 2005; Bartkova et al., 2005b, 2007; Fan et al., 2006; Tort et al., 2006; Nuciforo et al., 2007).

One study performed on a limited number of samples from MDS and AML patients has also suggested that γ-H2AX foci and the phosphorylation of ATM and Chk-2 is particularly high in the early stages of MDS with a statistically non-significant tendency to decrease in AML (Horibe et al., 2007). Driven by these premises, we decided to re-investigate the putative contribution of the endogenous DDR to MDS and AML on a panel of patient-derived cell lines and bone marrow biopsies.

Results and discussion

Heterogeneity in the radioresponse and radiosensitization of MDS/AML cell lines

Although MDS and AML are not treated by radiotherapy, ionizing irradiation provides a method to inflict a DDR that is not influenced by multiple drug resistance pumps or tumor cell metabolism. We irradiated a panel of distinct cell lines from patients with MDS or AML to comparatively study their radioresponse in vitro. The P39 cells are originally derived from a patient diagnosed with MDS (who developed AML subsequently) (Nagai et al., 1984). MOLM-13 represents a post-MDS AML, whereas MV4-11 and KG-1 were established from patients diagnosed with de novo AML (Koeffler and Golde, 1980; Lange et al., 1987; Matsuo et al., 1997). MV4-11 has a more mature (M5) phenotype, whereas KG-1 has an immature phenotype (Koeffler and Golde, 1980; Lange et al., 1987). The apoptotic response was monitored 24 h after γ-irradiation by assessing the dissipation of the mitochondrial transmembrane potential (ΔΨm, using the ΔΨm-sensitive dye DiOC6(3)) and the permeabilization of the plasma membrane (using the fluorochrome propidium iodine, PI). P39 and MOLM-13 cells were more radiosensitive than MV4-11 and KG-1 (Figures 1a and d). To determine the contribution of ATM and Chk-1 to the radioresponse of the cells, we inhibited ATM with KU-55933 (Hickson et al., 2004) and Chk-1 with UCN-01 (Graves et al., 2000). We found that ATM inhibition radiosensitized MV4-11 cells, re-establishing an apoptotic response that was similar to the radiosensitive P39 and MOLM-13 cells. However, KG-1 cells remained radioresistant, even in the presence of KU-55933 (Figures 1b and d). In contrast, doses of UCN-01 that efficiently inhibit Chk-1 (Graves et al., 2000) were capable of restoring the radiosensitivity of both MV4-11 and KG-1 cells (Figures 1c and d). We also monitored the cell cycle (by assessing PI incorporation into permeabilized cells) and found that KG-1 cells were particularly resistant against the radio-induced G2/M arrest (Figures 2a and d). In cells irradiated with 10 Gy, the ATM inhibitor KU-55933 reduced the radiation-induced G2/M arrest of all cell lines, yielding a relatively homogeneous pattern (Figures 2b and d). UCN-01 abolished the radiation-induced cell cycle arrest in P39 cells, reduced that of MOLM-13 cells and failed to affect MV4-11 and KG-1 cells (Figures 2c and d). These results indicate that distinct cell lines representing different degrees of MDS/AML greatly vary in their DDR as well as in their behavior with regard to ATM and Chk-1 inhibition.

Figure 1
figure1

Effect of γ-irradiation on apoptosis of a panel of MDS/AML cell lines. P39, MOLM-13, MV4-11 and KG-1 cells received the indicated dose of γ-irradiation and were cultured in the absence (a) or presence of KU-55933 (b) or 7-hydroxystaurosporine, also called UCN-01 (c), followed by the quantitation of the ΔΨm and the viability, as described in Materials and methods. Representative FACS pictograms are shown in (a–c) and quantitative data (means±s.d., triplicates) are provided in (d). This experiment was repeated four times, and data from one representative experiment are shown.

Figure 2
figure2

Effect of γ-irradiation on the cell cycle advancement of MDS/AML cell lines. Distinct cells were irradiated and were cultured in the absence (a) or presence of KU-55933 (b) or UCN-01 (c), followed by fixation, permeabilization and staining with PI. Representative FACS pictograms are shown in (a–c) and quantitative data (means±s.d., triplicates) are provided in (d). This experiment was repeated three times, and data from one representative experiment are shown.

To further refine this analysis, we determined the irradiation-induced cyclin B accumulation (which occurs in the G2 phase of the cell cycle) and histone H3 phosphorylation (which is mediated by cyclin B-dependent kinase-1 and affects chromatin during mitosis) comparatively in P39 and KG-1 cells. To assess the impact of irradiation on the cell cycle arrest in the absence of concomitant apoptosis, P39 and KG-1 cells were irradiated with 2 and 5 Gy. In both cell lines, irradiation caused an arrest in G2 (not M) 24 h post irradiation (Figure 3), but not at 1 h (Supplementary Figure 1). Once more, KG-1 cells required a higher dose of irradiation (5 Gy) than P39 cells to manifest this G2 blockade (Figure 3a). KU-55933 failed to abolish the G2 arrest in both cell lines and actually radiosensitized KG-1 cells, increasing the extent of G2-blockage in KG-1 cells to that observed in P39 cells (Figure 3b). Thus KG-1 cells blocked their cell cycle at 2 Gy in the presence (but not in the absence) of KU-55933 (Figure 3b). Caffeine, a combined ATM/ATR inhibitor, diminished the radioresponse of P39 cells, yet had no effect on KG-1 cells (Figure 3c). Altogether, these data provide evidence that cell lines derived from myeloid malignancies (MDS and AML) exhibit profound defects in their capability to mount an ATM-dependent checkpoint arrest. In addition, these results point to profound differences in the radioresponse of MDS (P39 cells)- and AML (KG-1 cells)- derived cell lines and demonstrate their varying degree to rely on ATM/ATR-mediated signaling.

Figure 3
figure3

Cyclin B accumulation and histone H3 phosphorylation in irradiated P39 and KG-1 cells. Twenty-four hours after irradiation with 0, 2 or 5 Gy the cells were subjected to a triple staining to determine DNA content (with DAPI), cyclin B expression and histone H3 phosphorylation. This was done for cells cultured in the absence (a) or presence of KU-55933 (b) or caffeine (c). Numbers refer to the percentage of cells found in the relevant quadrant or rectangle. Data are representative for one experiment out of three.

Spontaneous DDR in unirradiated AML, but not MDS cell lines

Driven by the unexpected heterogeneity in the ATM and Chk-1-modulated radiation response of distinct MDS/AML cell lines, we assessed the phosphorylation state of crucial components of the DDR pathway, before and after irradiation, in the absence or presence of ATM and Chk-1 inhibitors (Figure 1). KG-1 cells (and to a lower degree MV4-11 cells) show signs of a DDR (such as the activating phosphorylation of ATM on serine 1981) even in the absence of prior irradiation (Figure 4a). The activating phosphorylation of Chk-1 could be detected in MV4-11 cells before irradiation. Upon irradiation with 10 Gy, all cell lines except KG-1 manifested an increased phosphorylation of ATM, Chk-1 and H2AX. In accordance with the effects of 10 Gy on cell cycle advancement, P39 and MOLM-13 manifested a strong inhibitory phosphorylation of cyclin-dependent kinase-1 on tyrosine 15 and this phosphorylation event was less pronounced in 10 Gy-irradiated MV4-11 and KG-1 cells. As to be expected, KU-55933 reduced the phosphorylation of ATM (which is mainly an auto-phosporylation (Bakkenist and Kastan, 2003)) and reduced the phosphorylation of H2AX. UCN-01 abolished the phosphorylation of Chk-1 (but not that of ATM and H2AX) and reduced the difference in radio-induced cyclin-dependent kinase-1 phosphorylation among the different cell types (Figure 4a). To further explore the DDR of P39 and KG-1 cells, both cell lines were irradiated with comparatively lower doses (2 or 5 Gy) and the phosphorylation of critical DDR proteins was examined by immunoblot. Twenty-four hours post 5 Gy, P39 cells manifested a more pronounced Chk-2 phosphorylation than KG-1 cells, KU-55933 decreased this radiation-induced Chk-2 phosphorylation in both cell lines, whereas incubation with caffeine failed to subvert the phosphorylation of Chk-2 in both cell lines (Figure 4b), underscoring the perturbation of the ATM/Chk-2 axis in malignant myeloid cells.

Figure 4
figure4

Immunoblot detection of phorphorylation events linked to DDR. (a) The entire panel of MDS/AML cell lines was left untreated (in the logarithmic phase of growth) or γ-irradiated with 10 Gy and cultured in the absence (left panel) or presence of KU-55933 (middle panel) or UCN-01 (right panel), followed by lysis and immunoblot detection of the indicated phosphoproteins or proteins. All immunoblots shown for one single protein were performed simultaneously and revealed together, allowing for inter-panel comparisons. This experiment was repeated three times, yielding comparable results. (b) P39 and KG-1 cells were left untreated or γ-irradiated with 2 or 5 Gy and cultured in the absence (left panel) or presence of KU-55933 (middle panel) or caffeine (right panel), followed by immunoblot analyses as in (a).

Altogether, these results underscore the heterogeneity in the activation of the DDR among different MDS/AML cell lines, in accord with the heterogeneity in the apoptotic and cell cycle responses. Moreover, these results point to the presence of a constitutive (‘spontaneous’) DDR in KG-1 cells. To corroborate this conclusion, we performed three-color stainings of chromatin (stained with DAPI), phosphorylated Chk-1, Chk-2, ATM or γ-H2AX. Altogether, this technology corroborated the results obtained by immunoblotting in the sense that P39 cells more readily manifest the activating phosphorylation of DDR proteins than KG-1 cells after irradiation with 5 Gy (Figure 5d, Supplementary Figure 2). Although cells with a positive nuclear staining of phospho-Chk-1, phospho-Chk-2, phospho-ATM or γ-H2AX were rare among unirradiated P39 cells, 10% of unirradiated KG-1 stains exhibited a clear phosphorylation of ATM, Chk-1, Chk-2 and the presence of γ-H2AX in discrete nuclear speckles (‘DNA damage foci’) (Figures 5a and b, c). This spontaneous DDR of unirradiated KG-1 cells was inhibited by pretreatment with KU-55933, but not with caffeine (Figure 5c). Hence, MDS/AML cell lines manifest a variable degree of endogenous DDR. It is interesting to note that KG-1 cells, which exhibited signs of a spontaneous DDR, had a similar overall cell cycle distribution as P39 cells (Figure 2), which failed to exhibit signs of a spontaneous DDR. Moreover, the addition of KU-55933 did not affect the cell cycle of unirradiated KG-1 cells (Figures 2a and 3a). Finally, we found no clear correlation between the cell cycle advancement and ATM phosphorylation in KG-1 cells released from a double thymidine block (Supplementary Figure 3). Altogether, these findings suggest that the phosphorylation of ATM in unirradiated KG-1 cells did not inhibit cell cycle progression and hence was ‘uncoupled’ from its downstream consequences.

Figure 5
figure5

Immunofluorescence detection of DDR-linked phosphorylation events. Untreated P39 and KG-1 cells in the logarithmic phase of expansion were immobilized on polylysine slides and stained for the detection of the indicated phosphoproteins (a, b). This experiment was performed three times and the means±s.d. have been determined either for untreated cells (a, b, c) or for cells pretreated with KU-555933 or caffeine (c). In addition, the entire experiment was performed on irradiated cells (see also Supplementary Figure 2) and the percentage of cells expressing the indicated phosphoproteins was determined (d).

DDR in primary myeloblasts from MDS/AML patients

Bone marrow biopsies from patients with MDS or AML were subjected to the immunohistochemical detection of phospho-ATM, γ-H2AX (Figure 6, Table 1), phospho-Chk-1 or phospho-Chk-2 (Figure 7, Table 1). Bone marrow from healthy controls contained <1% of positive cells (Figures 6a and 7a), whereas bone marrow from MDS and AML patients exhibited a variable degree of positivity of phospho-ATM, γ-H2AX, phospho-Chk-1 or phospho-Chk-2 (Figures 6 and 7). The phosphorylation of ATM was relatively heterogeneous among all int-1 and int-2 MDS patients, yet increased in AML samples compared with high-risk MDS patients (Figure 6b), correlating significantly with the percentage of blasts in the bone marrow (Figure 6c). The presence of γ-H2AX+ positive cells was heterogeneous among all groups and, surprisingly, failed to increase in AML (Figure 6b) and to correlate with blast counts (Figure 6d) and with the extent of ATM phosphorylation (Figure 6e). When the pattern of Chk-1 and Chk-2 phosphorylation among different diagnostic groups was analysed, a radically different pattern emerged. Bone marrow from patients diagnosed with int-1 MDS contained as little phospho-Chk-1+ or phospho-Chk-2+ cells as bone marrow from healthy controls (Figures 7a and b), and the levels of Chk-1/Chk-2 phosphorylation increased only in int-2 and high-risk MDS patients, yet declined to near-to-background levels in bone marrow samples from AML patients (Figures 7a and b). There was no significant correlation between the percentage of malignant myeloblasts in the bone marrow and phospho-Chk-1+ (Figure 7c) or phospho-Chk-2+ cells (Figure 7d), whereas both phosphorylation events affecting the two different checkpoint kinases correlated among each other (Figure 7e). Moreover, we failed to detect any significant correlation between signs of the DDR and FAB subgroups of MDS/AML or between the pattern of DDR and alterations of the karyotype (not shown). Altogether, these results indicate that Chk-1 and Chk-2 phosphorylation may behave in accord with the paradigm established for solid tumors (that is upregulation during the pre-malignant stage but downregulation in fully transformed cells) whereas ATM is upregulated throughout and beyond transformation.

Figure 6
figure6

Immunohistochemical assessment of the phosphorylation of ATM and H2AX in bone marrow biopsies from control individuals and MDS/AML patients. Representative images from control individuals and patients with the indicated diagnoses are shown in (a) All information on patients can be retrieved from Table 1. (b) Box plot representation of the percentages of positive cells for the different subgroups of MDS (Int-1: intermediate-risk-1, Int-2: intermediate-risk-2, High: high-risk) and AML patients. (c) Correlation among the percentages of malignant bone marrow (BM) myeloblasts and phospho-ATM+ cells. (d) Correlation among the percentages of BM blasts and phospho-H2AX+ cells. (e) Correlation between phospho-ATM+ and phospho-H2AX+ cells.

Table 1 Characteristics of MDS and AML patients
Figure 7
figure7

Immunohistochemical detection of the phosphorylation of Chk-1 and Chk-2 in bone marrow biopsies from control individuals and MDS/AML patients. (a) Representative microphotographs (Int-1: intermediate-risk-1, Int-2: intermediate-risk-2, High: high-risk). (b) Box plot representation of the results obtained on the entire cohort of patients. (c) Correlation among the percentages of malignant bone marrow (BM) myeloblasts and phospho-Chk-1+ cells. (d) Relationship between BM blast and phospho-Chk-2+ cells. (e) Plot of the percentage of phospho-Chk-1+ versus phospho-Chk-2+ cells.

Concluding remarks

The assessment of signs of endogenous DDR in bone marrow biopsies from patients with MDS and AML as well as in patient-derived cell lines, revealed a major heterogeneity across (and sometimes correlating with) distinct diagnoses of MDS risk groups and AML. One particularly astonishing finding was that the activating (auto) phosphorylation of ATM did not correlate with the presence of clearly discernible γ-H2AX+ activation, suggesting that ATM may be inhibited in a fraction of patients, through a yet-to-be determined mechanism and/or that H2AX may be phosphorylated by other kinases than ATM (Fernandez-Capetillo et al., 2004; Thiriet and Hayes, 2005).

More importantly, our data point to an interruption of the DDR in AML that allows for the (auto) activation of ATM, yet causes the inhibition of Chk-1 and Chk-2. Thus, patients with AML manifested the activating phosphorylation of ATM, yet lacked any sign of Chk-1 or Chk-2 activation. The precise molecular mechanisms responsible for this ‘uncoupling’ effect remain elusive. An abundant literature documents that ATM can indirectly activate Chk-2, whereas Chk-1 is primarily activated by ATR, although some crosstalk exists between the ATM-Chk-2 and the ATR-Chk-1 axes (Shiloh, 2003; Bartek et al., 2007). Future work will have to assess how the functional link between activated ATM and downstream kinases such as Chk-1/Chk-2 is interrupted in AML. The in vitro characterization of a panel of cell lines revealed that MDS- as well as AML-derived cell lines exhibit profound defects in their capacity to mount a checkpoint response. Thus, the ATM-dependent checkpoint arrest is diminished in both cell lines, with the difference that MDS-derived P39 cells exhibited a later (most likely ATR-dependent) checkpoint response than KG-1 cells.

Accordingly, the ‘uncoupling’ detected in ex vivo patient samples was best recapitulated in unirradiated and irradiated KG-1 AML cells, as they exhibited a high level of ATM phosphorylation, yet an absent activation of Chk-1 and a diminished phosphorylation of Chk-2. Moreover, the inhibition of ATM failed to improve cell cycle progression of KG-1 cells, and the ATM phosphorylation could not be correlated with any kind of cell cycle blockade, indicating that the spontaneous phosphorylation of ATM, as observed in unirradiated KG-1 cells, is ‘uncoupled’ from its usual cell cycle-arresting and pro-apoptotic effects. Hence, KG-1 cells may be used for the further molecular characterization of the before mentioned ‘uncoupling’ effect. We found that ATM from KG-1 cells was enzymatically active, based on the fact that the direct ATM inhibitor KU-55933 (which targets the ATPase activity of ATM (Hickson et al., 2004)) abolishes the auto-phosphorylation of ATM and concomitantly inhibits H2AX phosphorylation. Hence, the presence, absence or mutation of other gene products must account for the ‘uncoupling’ effect. It should be noted that the ‘uncoupling’ effect is rather selective. Although activation of ATM appears to be uncoupled from the pro-apoptotic and cell cycle-arresting action (of Chk-1/Chk-2), the activation of ATM may remain coupled to antiapoptotic signals. Thus, we recently found that spontaneous or radiation-induced ATM activation can elicit the activation of the antiapoptotic transcription factor NF-κB, both in P39 and in KG-1 cells (Grosjean-Raillard et al., 2009).

The presence of activated ATM yet an invalidated DDR (and hence an open barrier against transformation/tumor progression) in AML raises two interesting questions. First, what is the cause of continued activation of ATM? It appears plausible, yet remains to be demonstrated, that continued oncogenic stress leading to prolonged DNA damage may explain the ATM activation. If so, the motor of ATM activation, endogenous DNA damage, may also constitute the motor of ongoing genomic instability and progressive leukemogenesis. Second, it has been suggested that ATM can mediate cytoprotective side effects in the response to DNA-damaging agents. Thus, ATM can stimulate the activation of the NF-κB pathway, leading to the expression of antiapoptotic and tumor-promoting gene products (Wu et al., 2006). Hence, it might be interesting to evaluate whether the continuous inhibition of ATM might accelerate or rather inhibit tumor progression in high-risk MDS and AML. We anticipate that responding to these questions may yield fundamental insights into the biology and perhaps into the clinical management of these myeloid malignancies.

Materials and methods

Patients

Samples of patients or healthy volunteers were assessed after obtaining informed consent and the diagnosis of AML and MDS was determined by cytology of peripheral blood and bone marrow, according to the WHO and FAB classification, as well as by conventional cytogenetic analysis. In patients with MDS, risk groups were determined using the International Prognostic Scoring System (Greenberg et al., 1997).

Cell lines and culture conditions

The P39/Tsugane cells were kindly provided by Dr Yoshida Takeda, Japan, whereas the KG-1 cells were a gift from Dr Martin Ruthardt, University of Frankfurt, Germany. MOLM-13 and MV4-11 cells were purchased from the Deutsche Sammlung von Mikroorganismen und Zellkulturen (DSMZ, Braunschweig, Germany). All cell lines were cultured in FCS-supplemented RPMI-1640 (Gibco, Gaihersburg, MD, USA). Unless specified differently, cells were seeded at concentrations of 1–2 × 105 cells/ml.

Assessment of apoptosis and cell cycle distribution

Cells (1 × 105) were suspended in 1 ml of culture medium and γ-irradiated with 0, 2, 5 and 10 Gray (Gy), respectively. The induction of apoptosis and cell cycle arrest were determined after 1 h and 24 h following irradiation. To assess the functionality of the G2/M checkpoint, cells were incubated in the absence and presence of pharmaceutical inhibitors of Chk-1 (UCN-01, 100 nm, National Cancer Institute, Bethesda, MD, USA), ATM (KU-55933, 10 μM, Calbiochem, San Diego, CA, USA) and caffeine (1 mM, Sigma, Steinheim, Germany).

Apoptotic cells were quantified by cytofluorometric analysis using a FACScan (Becton Dickinson, Mountain View, CA, USA) as described earlier (Castedo et al., 2002; Zamzami and Kroemer, 2004; Galluzzi et al., 2007). Thus, cells were stained with the vital dye propidium iodide (PI; 5 μg/ml; Sigma) and concomitantly with DiOC6(3) (3,3 dihexyloxacarbocyanine iodide; 40 nM; Molecular Probes, Eugene, OR, USA) for 15 min at 37 °C to determine the mitochondrial transmembrane potential.

For cell cycle analyses, cells were harvested, washed in PBS, stained with propidium iodide (PI, 25 μg/ml, Sigma) followed by an incubation period of 30 min at 37 °C as described earlier (Marzo et al., 1998; Zamzami et al., 2000). For analysis of cells in G2/M, ethanol-fixed cells were stained with an anticyclin B antibody (mouse monoclonal antibody, BD Biosciences, Franklin Lakes, NJ, USA) and an anti-phospho-histone-3-antibody (rabbit polyclonal antibody, Upstate, Lake Placid, NY, USA). Subsequently, cells were washed and incubated with the FITC- and PE-conjugated (FITC-conjugated goat anti-rabbit antibody obtained from Invitrogen, Carlsbad, CA, USA, PE-conjugated goat anti-mouse antibody purchased from BD Biosciences) secondary antibodies specific for mouse or rabbit IgG, and counterstained with DAPI (Invitrogen) to determine DNA_content. Cell cycle distribution was determined by cytofluorometric analysis using a FACS Vantage (Becton Dickinson).

Synchronization of cells by thymidine block

Cells in the logarithmic phase of growth were incubated overnight (16 h) with 2 mM thymidine (Sigma). Thymidine was removed the next morning by extensive washing, followed by a second cycle of overnight incubation/washing, followed by culture in complete medium to release the cells from cell cycle arrest as described earlier (Coquelle et al., 2006). Cells serving as a control were left without thymidine. Cells were fixed by ethanol after 0, 2, 4, 6, 8 and 10 h and cell cycle distribution was quantified by FACS analysis as described above.

Immunoblot analysis

Expression of DDR-associated proteins was evaluated at the indicated time points after γ-irradiation with 0, 2, 5 and 10 Gy. Lysates from 5 × 106 cells were separated on sodium dodecyl sulfate–polyacrylamide gels and electroblotted onto PVDF membranes following standard procedures (Boehrer et al., 2008). The membrane was blocked with 5% nonfat, dry milk and incubated with the respective primary antibody: Actin (mouse monoclonal antibody; Chemicon, Temecula, CA, USA), P-ATM-Ser1981 (mouse monoclonal antibody, Upstate), P-Chk-1-Ser317 (rabbit polyclonal antibody, Cell Signaling, Danvers, MA, USA), Chk-1 (rabbit polyclonal antibody, Santa Cruz Biotechnology, Santa Cruz, CA, USA), P-Chk-2-Ser68 (rabbit polyclonal antibody, Santa Cruz Biotechnology), γ-H2AX-Ser139 (mouse monoclonal antibody, Trevigen, Gaithersburg, MD, USA), P-Cdk1-Tyr15 (rabbit polyclonal antibody, Cell Signaling), or PCNA (mouse polyclonal antibody, BD Biosciences). Blots were stained with either goat anti-rabbit peroxidase-conjugated or goat anti-mouse peroxidase-conjugated secondary antibody (Amersham, Arlington Heights, IL, USA) and were revealed using an enhanced chemiluminescence detection system (Amersham).

Immunofluorescence

Slides for immunoflourescence were prepared 24 h after γ-irradiation with 0, 5 and 10 Gray as described earlier (Braun et al., 2006). Shortly, cells were allowed to adhere on polylysine-L coverslips (Sigma) and were fixed in 4% paraformaldehyde at room temperature. Cells were then permeabilized with sodium dodecyl sulfate 0.1% for 10 min, washed in PBS and stained with the indicated primary antibodies: P-ATM-Ser1981 (mouse monoclonal antibody, Upstate), P-Chk-1-Ser317 (rabbit polyclonal antibody, Cell Signaling), P-Chk-2-Ser68 (rabbit polyclonal antibody, Santa Cruz Biotechnology) and γ-H2AX-Ser139 (mouse monoclonal antibody, Trevigen). Antigens were revealed with the adequate secondary antibody coupled with Alexa 568 (red) or Alexa 488 (green) fluorochromes (Molecular Probes/Invitrogen). DNA of cells was counterstained with Hoechst 33324 or DAPI (Molecular Probes/Invitrogen). Two hundred cells for each slide were examined and counted independently with a LSM 510 confocal microscope (Zeiss, Thornwood, NY, USA) at × 63 magnification. Background correction of fluorescence was performed with the LSM 5 image browser (Zeiss).

Immunohistochemistry

Paraffin-mounted bone marrow core biopsy sections were deparaffinized and hydrated antigens were retrieved. Specimens were exposed to 10 mM citrate buffer (pH 6.0) and heated for 30 min in a water bath. Subsequently tumor sections were incubated for 60 min with the respective antibodies: P-ATM-Ser1981 (mouse monoclonal antibody, Rockland, PA, USA), P-Chk1-Ser317 (rabbit polyclonal antibody, Cell Signaling), P-Chk-2-Ser68 (rabbit polyclonal antibody, Santa Cruz Biotechnology) and γ-H2AX-Ser139 (mouse monoclonal antibody, Upstate). Antibody binding was detected using the ABC kit with NovaRED as the substrate (Vectastain Elite, Vector Laboratories, Burlingame, CA, USA) and slides were counterstained by Mayer's hematoxylin (Sigma). Expression of DDR proteins was determined by screening the whole area of the bone marrow biopsy using a × 10 magnification of a LSM 510 confocal microscope (Zeiss). Three chambers representative of the slide were examined at a magnification of × 40, and the percentage of positively stained cells was assessed by counting at least 200 cells.

Statistics

Statistics were calculated with the help of Excel Software (Microsoft, Redmond, WA, USA), SPSS software (SAS Institute, Cary, NC, USA) and Scion Image 4.0 (Scion Corporation, Frederick, MD, USA). Expression of DDR proteins evaluated in bone marrow biopsies of MDS and AML patients are depicted using a box plot graph, where the horizontal represents the median, the box the 25th percentile and the whiskers the extreme. Statistical significance was determined using the Levene test assessing the equality of the variances. Correlation between the expression of individual DDR proteins and the percentage of bone marrow blast cells is depicted using the Pearson index.

Conflict of interest

The authors declare no conflict of interest.

References

  1. Bakkenist CJ, Kastan MB . (2003). DNA damage activates ATM through intermolecular autophosphorylation and dimer dissociation. Nature 421: 499–506.

  2. Bartek J, Bartkova J, Lukas J . (2007). DNA damage signalling guards against activated oncogenes and tumour progression. Oncogene 26: 7773–77739.

  3. Bartkova J, Bakkenist CJ, Rajpert-De Meyts EA et al. (2005a). ATM activation in normal human tissues and testicular cancer. Cell Cycle 4: 838–845.

  4. Bartkova J, Horejsi Z, Koed K, Kramer A, Tort F, Zieger K et al. (2005b). DNA damage response as a candidate anti-cancer barrier in early human tumorigenesis. Nature 434: 864–870.

  5. Bartkova J, Horejsi Z, Sehested M, Nesland JM, Rajpert-De Meyts E, Skakkebaek NE et al. (2007). DNA damage response mediators MDC1 and 53BP1: constitutive activation and aberrant loss in breast and lung cancer, but not in testicular germ cell tumours. Oncogene 26: 7414–7422.

  6. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galton DA, Gralnick HR et al. (1982). Proposals for the classification of the myelodysplastic syndromes. Br J Haematol 51: 189–199.

  7. Boehrer S, Adès L, Braun T, Galluzzi L, Grosjean J, Fabre C et al. (2008). Erlotinib exhibits antineoplastic off-target effects in AML and MDS: a preclinical study. Blood 111: 2170–2780.

  8. Braun T, Carvalho G, Coquelle A, Vozenin MC, Lepelley P, Hirsch F et al. (2006). NF-kappaB constitutes a potential therapeutic target in high-risk myelodysplastic syndrome. Blood 107: 1156–1165.

  9. Castedo M, Ferri K, Roumier T, Metivier D, Zamzami N, Kroemer G . (2002). Quantitation of mitochondrial alterations associated with apoptosis. J Immunol Methods 265: 39–47.

  10. Coquelle A, Mouhamad S, Pequignot MO, Braun T, Carvalho G, Vivet S et al. (2006). Enrichment of non-synchronized cells in the G1, S and G2 phases of the cell cycle for the study of apoptosis. Biochem Pharmacol 72: 1396–1404.

  11. DiTullio Jr RA, Mochan TA, Venere M, Bartkova J, Sehested M, Bartek J et al. (2002). 53BP1 functions in an ATM-dependent checkpoint pathway that is constitutively activated in human cancer. Nat Cell Biol 4: 998–1002.

  12. Fan C, Quan R, Feng X, Gillis A, He L, Matsumoto ED et al. (2006). ATM activation is accompanied with earlier stages of prostate tumorigenesis. Biochim Biophys Acta 1763: 1090–1097.

  13. Fernandez-Capetillo O, Lee A, Nussenzweig M, Nussenzweig A . (2004). H2AX: the histone guardian of the genome. DNA Repair (Amst) 3: 959–967.

  14. Galluzzi L, Maiuri MC, Vitale I, Zischka H, Castedo M, Zitvogel L et al. (2007). Cell death modalities: classification and pathophysiological implications. Cell Death Differ 14: 1237–1243.

  15. Gorgoulis VG, Vassiliou LV, Karakaidos P, Zacharatos P, Kotsinas A, Liloglou T et al. (2005). Activation of the DNA damage checkpoint and genomic instability in human precancerous lesions. Nature 434: 907–913.

  16. Graves PR, Yu L, Schwarz JK, Gales J, Sausville EA, O'Connor PM et al. (2000). The Chk1 protein kinase and the Cdc25C regulatory pathways are targets of the anticancer agent UCN-01. J Biol Chem 275: 5600–5605.

  17. Greenberg P, Cox C, LeBeau MM, Fenaux P, Morel P, Sanz G et al. (1997). International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood 89: 2079–2088.

  18. Grimwade D . (2001). The clinical significance of cytogenetic abnormalities in acute myeloid leukaemia. Best Pract Res Clin Haematol 14: 497–529.

  19. Grimwade D, Moorman A, Hills R, Wheatley K, Walker H, Harrison G et al. (2004). Impact of karyotype on treatment outcome in acute myeloid leukemia. Ann Hematol 83 (Suppl 1): S45–S48.

  20. Grimwade D, Walker H, Harrison G, Oliver F, Chatters S, Harrison CJ et al. (2001). The predictive value of hierarchical cytogenetic classification in older adults with acute myeloid leukemia (AML): analysis of 1065 patients entered into the United Kingdom Medical Research Council AML11 trial. Blood 98: 1312–1320.

  21. Grosjean-Raillard J, Tailler M, Adès L, Perfettini JL, Fabre C, Braun T et al. (2009). ATM mediates constitutive NF-kappaB activation in high-risk myelodysplastic syndrome and acute myeloid leukemia. Oncogene 28: 1099–1109.

  22. Hickson I, Zhao Y, Richardson CJ, Green SJ, Martin NM, Orr AI et al. (2004). Identification and characterization of a novel and specific inhibitor of the ataxia-telangiectasia mutated kinase ATM. Cancer Res 64: 9152–9159.

  23. Horibe S, Takagi M, Unno J, Nagasawa M, Morio T, Arai A et al. (2007). DNA damage check points prevent leukemic transformation in myelodysplastic syndrome. Leukemia 21: 2195–2198.

  24. Kastan MB, Bartek J . (2004). Cell-cycle checkpoints and cancer. Nature 432: 316–323.

  25. Klaus M, Haferlach T, Schnittger S, Kern W, Hiddemann W, Schoch C . (2004). Cytogenetic profile in de novo acute myeloid leukemia with FAB subtypes M0, M1, and M2: a study based on 652 cases analyzed with morphology, cytogenetics, and fluorescence in situ hybridization. Cancer Genet Cytogenet 155: 47–56.

  26. Koeffler HP, Golde DW . (1980). Human myeloid leukemia cell lines: a review. Blood 56: 344–350.

  27. Lange B, Valtieri M, Santoli D, Caracciolo D, Mavilio F, Gemperlein I et al. (1987). Growth factor requirements of childhood acute leukemia: establishment of GM-CSF-dependent cell lines. Blood 70: 192–199.

  28. Marzo I, Susin SA, Petit PX, Ravagnan L, Brenner C, Larochette N et al. (1998). Caspases disrupt mitochondrial membrane barrier function. FEBS Lett 427: 198–202.

  29. Matsuo Y, MacLeod RA, Uphoff CC, Drexler HG, Nishizaki C, Katayama Y et al. (1997). Two acute monocytic leukemia (AML-M5a) cell lines (MOLM-13 and MOLM-14) with interclonal phenotypic heterogeneity showing MLL-AF9 fusion resulting from an occult chromosome insertion, ins(11;9)(q23;p22p23). Leukemia 11: 1469–1477.

  30. Nagai M, Seki S, Kitahara T, Abe T, Minato K, Watanabe S et al. (1984). A novel human myelomonocytoid cell line, P39/Tsugane, derived from overt leukemia following myelodysplastic syndrome. Gann 75: 1100–1107.

  31. Nuciforo PG, Luise C, Capra M, Pelosi G, d'Adda di Fagagna F . (2007). Complex engagement of DNA damage response pathways in human cancer and in lung tumor progression. Carcinogenesis 28: 2082–2088.

  32. Parker JE, Mufti GJ . (1998). Ineffective haemopoiesis and apoptosis in myelodysplastic syndromes. Br J Haematol 101: 220–230.

  33. Parker JE, Mufti GJ . (2000). Excessive apoptosis in low risk myelodysplastic syndromes (MDS). Leuk Lymphoma 40: 1–24.

  34. Parker JE, Mufti GJ . (2001). The role of apoptosis in the pathogenesis of the myelodysplastic syndromes. Int J Hematol 73: 416–428.

  35. Shiloh Y . (2003). ATM and related protein kinases: safeguarding genome integrity. Nat Rev Cancer 3: 155–168.

  36. Slovak ML, Kopecky KJ, Cassileth PA, Harrington DH, Theil KS, Mohamed A et al. (2000). Karyotypic analysis predicts outcome of preremission and postremission therapy in adult acute myeloid leukemia: a Southwest Oncology Group/Eastern Cooperative Oncology Group Study. Blood 96: 4075–4083.

  37. Thiriet C, Hayes JJ . (2005). Chromatin in need of a fix: phosphorylation of H2AX connects chromatin to DNA repair. Mol Cell 18: 617–622.

  38. Tort F, Bartkova J, Sehested M, Orntoft T, Lukas J, Bartek J . (2006). Retinoblastoma pathway defects show differential ability to activate the constitutive DNA damage response in human tumorigenesis. Cancer Res 66: 10258–10263.

  39. Vardiman JW, Harris NL, Brunning RD . (2002). The World Health Organization (WHO) classification of the myeloid neoplasms. Blood 100: 2292–2302.

  40. Wu ZH, Shi Y, Tibbetts RS, Miyamoto S . (2006). Molecular linkage between the kinase ATM and NF-kappaB signaling in response to genotoxic stimuli. Science 311: 1141–1146.

  41. Zamzami N, Kroemer G . (2004). Methods to measure membrane potential and permeability transition in the mitochondria during apoptosis. Methods Mol Biol 282: 103–115.

  42. Zamzami N, El Hamel C, Maisse C, Brenner C, Muñoz-Pinedo C, Belzacq AS et al. (2000). Bid acts on the permeability transition pore complex to induce apoptosis. Oncogene 19: 6342–6350.

Download references

Acknowledgements

SB and LA contributed equally to this paper. SB receives a scholarship from the Deutsche Forschungsgemeinschaft, LA receives a scholarship from Assistance Publique-Hopitaux de Paris and Caisse Nationale d'Assurance Maladie des Professions Indépendantes. GK is supported by Cancéropôle Ile-de-France, Institut National du Cancer, Fondation de France, Association Laurette Fugain, Cent pour Sang la Vie, Agence National de la Recherche and the European Commission (ApoSys, ChemoRes., Apopt-Train, RIGHT). The project was carried out with the support from the Gutermuth-Stiftung. SB, LA, NT, LG, CF, MT and KAO. performed and analysed the experiments. WKH., SK., GB, OGO., MR, CG, VE, SdeB., ST and PF provided bone marrow biopsies and patient data. SB and GK designed the study and wrote the paper.

Author information

Correspondence to G Kroemer.

Additional information

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc).

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Boehrer, S., Adès, L., Tajeddine, N. et al. Suppression of the DNA damage response in acute myeloid leukemia versus myelodysplastic syndrome. Oncogene 28, 2205–2218 (2009) doi:10.1038/onc.2009.69

Download citation

Keywords

  • leukemogenesis
  • apoptosis
  • cell cycle
  • ATM
  • Chk-1

Further reading