DNA-damaging therapy is the basis for treatment of most cancers, including B-cell precursor acute lymphoblastic leukemia (BCP-ALL, hereafter ALL). We have previously shown that cAMP-activating factors present in the bone marrow render ALL cells less sensitive to DNA damage–induced apoptosis, by enhancing autophagy and suppressing p53. To sensitize ALL cells to DNA-damaging therapy, we have searched for novel targets that may counteract the effects induced by cAMP signaling. In the current study, we have identified PARP1 as a potential target. We show that the PARP1 inhibitors olaparib or PJ34 inhibit cAMP-mediated autophagy and thereby potentiate the DNA-damaging treatment. Furthermore, we reveal that cAMP-mediated PARP1 activation is preceded by induction of reactive oxygen species (ROS) and results in depletion of nicotinamide adenine dinucleotide (NAD), both of which are autophagy-promoting events. Accordingly, we demonstrate that scavenging ROS by N-acetylcysteine and repleting NAD independently reduce DNA damage-induced autophagy. In addition, olaparib augmented the effect of DNA-damaging treatment in a human xenograft model of ALL in NOD-scidIL2Rgammanull mice. On the basis of the current findings, we suggest that PARP1 inhibitors may enhance the efficiency of conventional genotoxic therapies and thereby provide a novel treatment strategy for pediatric patients with ALL.

Implications:

PARP1 inhibitors augment the DNA damage–induced killing of ALL cells by limiting the opposing effects of cAMP-mediated autophagy, which involves ROS-induced PARP1 activation and depletion of cellular NAD levels.

B-cell precursor acute lymphoblastic leukemia (BCP-ALL, hereafter ALL) is the most common type of cancer diagnosed in children today (1). Steadily improved multimodal chemotherapy regimens over the last decades have caused the survival rate of this disease to reach nearly 90%. Despite this high survival rate, ALL is still the most common cause of cancer-related deaths in children, and many survivors struggle with severe long-term morbidity due to treatment-related sequelae (2, 3). Continued research on ALL is therefore needed to improve the treatment outcome of this disease and to diminish the severe side effects of current therapies.

Previous research from our group has shown that activation of the cyclic adenosine monophosphate (cAMP)/protein kinase A signaling pathway reduces DNA damage–mediated killing of ALL cells (4–8). ALL develops within stem cell niches in the bone marrow in close proximity with stromal cells. Bone marrow–derived stromal cells are known to secrete signaling molecules that may promote tumor growth and lead to treatment resistance (9–11). One of the secreted factors is prostaglandin E2 (PGE2) which in ALL cells stimulates formation of endogenous cAMP through ligation of prostanoid receptor 2 and activation of adenylyl cyclase (12, 13). Previous work from our group has suggested that the secretion of PGE2 from bone marrow stromal cells might be a driving force in development of ALL and reduce the effect of DNA-damaging therapy (6). We showed that activation of the cAMP signaling pathway in ALL cells inhibits p53-dependent DNA damage–induced cell death concomitant with induced activation of p53-independent autophagy (4–8, 14). Furthermore, we demonstrated that inhibiting PGE2 levels by indomethacin slowed down the leukemia progression in a human xenograft model of ALL (14), suggesting that the PGE2-cAMP signaling pathway may be a driving force in ALL development.

One of the key enzymes activated in response to single-strand DNA damage, is PARP1. PARP1 is encoded by the PARP1 gene in humans, and is responsible for poly ADP-ribosylation (parylation) of nuclear proteins (15). The process involves covalently attaching ADP-ribose chains both to PARP1 itself, and to other proteins such as histones and DNA-associated proteins. Parylation is involved in regulation of a multitude of cellular processes, such as chromatin organization, single-stranded DNA repair, transcription, as well as replication (16–18). In addition, PARP1 has been implicated in the induction of autophagy (19–21).

In the current study, we address the mechanisms whereby cAMP signaling enhances autophagy and thereby inhibits DNA damage–induced killing of ALL cells. Our data suggest that PARP1 may be a key factor involved in cAMP-mediated resistance to conventional genotoxic chemotherapy of ALL. Hence, we demonstrate that PARP1 inhibtors counteract the cAMP-mediated enhancement of DNA damage–induced autophagy and thereby enhances the cell death. We identify both upstream and downstream mediators of PARP1 activation, and we demonstrate that inhibition of PARP1 augments the ability of DNA-damaging treatment to delay the progression of leukemia in a human xenograft model of ALL in NOD-scidIL2Rgammanull (NSG) mice. Taken together, our results suggest that inhibition of PARP1 may enhance the efficiency of conventional therapy of ALL.

Cell culturing and primary cell isolation

The REH cell line is derived from a patient with BCP-ALL (22) and was purchased from the ATCC. The cells were maintained at a density between 2 × 105 and 1 × 106 cells per milliliter in RPMI1640 (Lonza) supplemented with 10% heat-inactivated FBS, 125 U/mL penicillin, and 125 μg/mL streptomycin. Cells were routinely tested for Mycoplasma (GATC Biotech).

To verify the physiologic relevance of the data, we obtained on REH cells, key experiments were repeated on primary leukemic blasts derived from newly diagnosed patients with ALL. Because of restricted number of cells obtained from the bone marrow aspirates, selected experiments were performed. The ALL blasts were isolated from bone marrow aspirates as described previously (4). The proportion of ALL blasts within the sample was determined by flow cytometry using antibodies against CD19 (MACS Miltenyi Biotec #130-091-328) and CD10 (BioLegend #312218, BioLegend). Characteristics for each patient included in the study are presented in Table 1. The REH cells as well as all the patient samples included in the study were p53 wild type. The collection of bone marrow aspirates was performed after written informed consent by parents, in accordance with the Declaration of Helsinki. The collection of samples is approved by the Regional Ethics Committee of Norway region South-East C (REK 2014/883).

Table 1.

Patient characteristics.

ALL#44ALL#45ALL#65ALL#71ALL#75
Age, years 10 11 
Sex 
Bone marrow infiltration at diagnosis (% CD19+/CD10+73 90 90 81 85 
Cytogenetics Trisomy 21 Del9p21 Hyperdiploidy Hyperdiploidy iAMP21 
ALL#44ALL#45ALL#65ALL#71ALL#75
Age, years 10 11 
Sex 
Bone marrow infiltration at diagnosis (% CD19+/CD10+73 90 90 81 85 
Cytogenetics Trisomy 21 Del9p21 Hyperdiploidy Hyperdiploidy iAMP21 

DNA damage induction by irradiation or doxorubicin

DNA damage was induced in the ALL cells by either treating the cells with X-ray irradiation (IR) using an X-Strahl RS320 X-ray irradiator at a rate of 3.9 Gy/minute or by doxorubicin (Sigma-Aldrich). The two approaches generally resulted in the same extent of DNA damage–induced cell death as measured by flow scatter analysis. In certain assays such as analysis of propidium iodide (PI) and CYTO-ID, doxorubicin interfered with the detection of these fluorochromes. For this reason, we predominantly used IR as a source to introduce DNA damage in the ALL cells. For short-term experiments (≤ 30 minutes), it was generally more practical to use doxorubicin than IR, and doxorubicin was therefore frequently used in experiments such as analysis of parylation.

Flow cytometry

All flow cytometry analyses were performed on a NovoCyte (Acea Biosciences Inc.) equipped with three lasers (488, 405, 640 nm) and 13 detection channels. For cell death analyses, cells were incubated with PI (20 μg/mL) for 10 minutes at 4°C and detected in the 572/28 channel. Changes in the mitochondrial membrane potential were determined by incubating the cells with JC-1 (15 μg/mL) at 37°C for 15 minutes before analysis. Staining of autophagosomes was performed by using the CYTO-ID Autophagy detection kit (Enzo Life Sciences) according to manufacturer's protocol and detected in the 530/30 channel. Levels of reactive oxygen species (ROS) were measured by using the CellROX Green Oxidative Stress Reagent (Thermo Fisher Scientific). The cells were incubated with CellROX (5 μmol/L) for 30 minutes at 37°C and detected in the 530/30 channel. CellROX and CYTO-ID data are presented as mean fluorescence intensity (MFI) emitted by the dye.

Statistical analyses

Statistical analyses were performed by doing the paired two-tailed Student t test using the GraphPad Prism7 software (GraphPad Software Inc.). Groups were found to be significantly different from each other when the P value was below 0.05. Graphs are presented as mean values from at least three independent experiments as specified in figure legends. Error bars indicate the SEM.

Data availability

All data generated or analyzed during this study are included in the published article (and its Supplementary Data files). Any materials may be made available from the corresponding author on reasonable request.

For further details about reagents, antibodies, and methods included in the study, see Supplementary Materials and Methods.

Inhibition of PARP1 abolishes cAMP-mediated induction of autophagy

We recently showed that autophagy has a critical role in the cAMP-mediated inhibition of DNA damage–induced apoptosis in ALL cell lines and primary leukemic cells (8). In light of the previously suggested role of protein parylation in autophagy (19–21), we here addressed the involvement of PARP1 in cAMP-mediated induction of autophagy in ALL cells. To this end, ALL-derived REH cells were treated with PARP1 inhibitors olaparib or PJ34, prior to exposing the cells to the adenylyl cyclase activator forskolin and IR. Autophagy was monitored by detecting the lipidated form of LC3 (LC3-II) by Western blot analyses (23), or by measuring CYTO-ID intensity by flow cytometry. The CYTO-ID dye is known to stain autophagic vesicles (24), and we have previously demonstrated similar results obtained by analyses of CYTO-ID staining and by detection of LC3-II (8). Forskolin enhanced the IR-induced autophagy, as revealed both by CYTO-ID staining and LC3-II detection (Fig. 1AD). Olaparib and PJ34 reduced the forskolin/IR-induced LC3-II levels by 75% and 37%, respectively, and the inhibiting effects of the two PARP1 inhibitors were confirmed by CYTO-ID staining (Fig. 1AD). Importantly, olaparib also inhibited autophagy induced by forskolin and IR in primary leukemic cells from 4 patients with ALL (Fig. 1E, see Table 1 for patient characteristics). Taken together, these results support the role of PARP1 in autophagy induced both by cAMP signaling and DNA damage.

Figure 1.

Inhibition of PARP1 abolishes cAMP-mediated induction of autophagy. REH cells (6 × 105 cells/mL) were incubated in the presence or absence of olaparib (10 μmol/L, A) or with PJ34 (20 μmol/L, B) for 30 minutes, followed by treatment with or without forskolin (forsk, 60 μmol/L) for 45 minutes prior to IR (10 Gy). BafA1 (2 nmol/L) was added to all samples 45 minutes before IR. The samples were harvested 6 hours after IR, and total lysates were subjected to immunoblot analyses with antibodies against LC3 and Calnexin (CANX). Left, One representative Western blot analysis of five is shown. Right, Ratios of LC3-II signal intensity relative to CANX signal, normalized to the ratio in untreated (Ctrl) cells. REH cells (4 × 105 cells/mL) were treated in the presence or absence of olaparib (10 μmol/L, C) or PJ34 (20 μmol/L, D) in the presence or absence of forsk, IR and BafA1 as described in A. Autophagy was determined 6 hours after IR by CYTO-ID staining, and the results are presented relative to the level in untreated (Ctrl) cells. The data represent the mean ± SEM, n = 3. *, P < 0.05; **, P < 0.01; §, P < 0.05; §§, P < 0.01 relative to sample treated without PARP1 inhibitor (paired t test). E, ALL cells (5 × 105 cells) from 4 different patients with ALL were treated with or without olaparib (5 μmol/L) for 30 minutes, followed by treatment with or without forskolin (60 μmol/L) for 45 minutes prior to IR (5 Gy). Cells were harvested 6 hours after IR and stained with CYTO-ID. Each symbol represents 1 patient.

Figure 1.

Inhibition of PARP1 abolishes cAMP-mediated induction of autophagy. REH cells (6 × 105 cells/mL) were incubated in the presence or absence of olaparib (10 μmol/L, A) or with PJ34 (20 μmol/L, B) for 30 minutes, followed by treatment with or without forskolin (forsk, 60 μmol/L) for 45 minutes prior to IR (10 Gy). BafA1 (2 nmol/L) was added to all samples 45 minutes before IR. The samples were harvested 6 hours after IR, and total lysates were subjected to immunoblot analyses with antibodies against LC3 and Calnexin (CANX). Left, One representative Western blot analysis of five is shown. Right, Ratios of LC3-II signal intensity relative to CANX signal, normalized to the ratio in untreated (Ctrl) cells. REH cells (4 × 105 cells/mL) were treated in the presence or absence of olaparib (10 μmol/L, C) or PJ34 (20 μmol/L, D) in the presence or absence of forsk, IR and BafA1 as described in A. Autophagy was determined 6 hours after IR by CYTO-ID staining, and the results are presented relative to the level in untreated (Ctrl) cells. The data represent the mean ± SEM, n = 3. *, P < 0.05; **, P < 0.01; §, P < 0.05; §§, P < 0.01 relative to sample treated without PARP1 inhibitor (paired t test). E, ALL cells (5 × 105 cells) from 4 different patients with ALL were treated with or without olaparib (5 μmol/L) for 30 minutes, followed by treatment with or without forskolin (60 μmol/L) for 45 minutes prior to IR (5 Gy). Cells were harvested 6 hours after IR and stained with CYTO-ID. Each symbol represents 1 patient.

Close modal

Inhibition of PARP1 reduces cAMP-mediated survival of IR-treated ALL cells

Having now established a role of PARP1 in cAMP-mediated autophagy, we next explored its potential function in the cAMP-mediated inhibition of DNA damage–induced cell death. REH cells were pretreated with the PARP1 inhibitors olaparib or PJ34 at increasing concentrations, followed by exposure to forskolin in the presence or absence of IR. Cell death was measured by PI staining 24 hours after IR. As shown in Fig. 2, both olaparib and PJ34 reversed the protective function of forskolin in IR-mediated cell death. Olaparib at 10 μmol/L reduced forskolin-mediated protection from 40% to 8% (Fig. 2A), whereas PJ34 at 10 μmol/L nearly completely reversed the protective effects of forskolin on IR-induced cell death from 44% to 5% (Fig. 2B). A third PARP inhibitor, veliparib, reduced the forskolin-mediated inhibition of DNA damage–induced cell death from 43% to 15% (Supplementary Fig. S1), The effects of the PARP inhibitors on DNA damage–induced cell death was confirmed by staining the cells with the cyanine dye JC-1 to measure apoptotic loss of the mitochondrial membrane potential. The dye accumulates in the mitochondria and forms red aggregates in viable cells, and green monomers in apoptotic cells (25). Figure 2C shows the scatter plots of one representative experiment of REH cells pretreated with olaparib, followed by incubation with forskolin and IR. The monomers are circled in red, and as shown in Fig. 2D, olaparib reduced the inhibitory effect of forskolin on IR-induced JC-1 monomers from 41% to 20%. We therefore concluded that PARP1 is involved in the protective effects of cAMP signaling on DNA damage–induced cell death, and consequently that PARP1 inhibitors are able to enhance the killing of ALL cells.

Figure 2.

Inhibition of PARP1 counters cAMP-mediated protection of DNA damage–induced cell death. REH cells (3 × 105 cells/mL) were incubated in the presence or absence of olaparib (A) or PJ34 (B) at the indicated concentrations for 30 minutes, followed by incubation with or without forskolin (forsk, 60 μmol/L) for 45 minutes prior to IR (10 Gy). PI staining was performed 24 hours after IR, and the percentage of PI-positive cells was analyzed by flow cytometry. The results are presented as mean ± SEM, n = 5. *, P < 0.05; **, P < 0.01 (paired t test). C and D, REH cells (4 × 106 cells/mL) were treated with olaparib (10 μmol/L), forsk and IR as described in A. JC-1 staining was performed 24 hours after IR as described in Materials and Methods. C, Scatter analysis of JC-1 monomer formation (indicated by red circles) by flow cytometry in one representative experiment. Debris were excluded in the analysis. Excitation wavelength: 405 nm. Detection channels: 530/30 and 572/28. D, Percentage of cells that have formed JC-1 monomers. The data represent the mean ± SEM, n = 4. *, P < 0.05 (paired t test). E, REH cells (4 × 106 cells/mL) were treated with forsk and IR as described in A. Olaparib was added 2 hours after IR at the indicated concentrations. The data represent the mean ± SEM, n = 4. *, P < 0.05 (paired t test).

Figure 2.

Inhibition of PARP1 counters cAMP-mediated protection of DNA damage–induced cell death. REH cells (3 × 105 cells/mL) were incubated in the presence or absence of olaparib (A) or PJ34 (B) at the indicated concentrations for 30 minutes, followed by incubation with or without forskolin (forsk, 60 μmol/L) for 45 minutes prior to IR (10 Gy). PI staining was performed 24 hours after IR, and the percentage of PI-positive cells was analyzed by flow cytometry. The results are presented as mean ± SEM, n = 5. *, P < 0.05; **, P < 0.01 (paired t test). C and D, REH cells (4 × 106 cells/mL) were treated with olaparib (10 μmol/L), forsk and IR as described in A. JC-1 staining was performed 24 hours after IR as described in Materials and Methods. C, Scatter analysis of JC-1 monomer formation (indicated by red circles) by flow cytometry in one representative experiment. Debris were excluded in the analysis. Excitation wavelength: 405 nm. Detection channels: 530/30 and 572/28. D, Percentage of cells that have formed JC-1 monomers. The data represent the mean ± SEM, n = 4. *, P < 0.05 (paired t test). E, REH cells (4 × 106 cells/mL) were treated with forsk and IR as described in A. Olaparib was added 2 hours after IR at the indicated concentrations. The data represent the mean ± SEM, n = 4. *, P < 0.05 (paired t test).

Close modal

In a clinical setting, the leukemic cells in patients would already be exposed to cAMP-stimulating factors in the bone marrow at the time of any treatment (12, 13). It was therefore encouraging to find that olaparib added 2 hours after exposure to forskolin and IR was still able to reverse the cAMP-mediated inhibition of DNA damage–induced cell death (Fig. 2E).

cAMP signaling induces parylation in ALL cells

PARP1 carries out its functions primarily via parylation of target proteins (15), and we therefore investigated whether cAMP signaling enhances the parylation process in DNA damage–treated ALL cells. PARP1 is usually the most frequently parylated protein by itself (26–28), and as shown by Western blot analyses in Fig. 3A, the parylation of PARP1 was transiently induced in REH cells by IR and cAMP signaling. Parylated PARP1 was noted as early as 2 minutes after IR and 5 minutes after addition of forskolin. The parylation was no longer detectable after 30 minutes, most likely due to par-degrading enzymes (29, 30). To verify the enhanced parylation of PARP1, we reprobed the blot with a PARP1 antibody (clone 46D11) that binds to the region of PARP1 that undergoes allosteric rearrangement upon DNA binding and subsequent activation (31). The signal obtained by the clone 46D11 antibody inversely correlated with the PAR signal (Fig. 3A), suggesting that the loss in signal may be due to activation of PARP1. The presence of equal levels of total PARP1 in all samples was revealed by employing the polyclonal PARP1 antibody (Fig. 3A). Figure 3B shows the signal intensity of parylated PARP1 relative to the loading control (GAPDH) 5 minutes after IR. Forskolin treatment increased the level of parylated PARP1 2-fold relative to the control, and had an additive effect on the IR-mediated parylation. The ability of PARP1 inhibitors to prevent parylation was proven by pretreating the REH cells with olaparib for 30 minutes prior to IR and forskolin treatment (Fig. 3C).

Figure 3.

cAMP signaling and IR induces parylation in REH cells. A, REH cells (1 × 106 cells/mL) were incubated in the presence or absence of forskolin (forsk, 60 μmol/L) and IR (10 Gy), and the cells were harvested at the indicated timepoints. Total lysates were subjected to immunoblot analyses with antibodies against PAR and GAPDH. One representative Western blot analysis of seven is shown. B, The cells were treated as described in A and harvested 5 minutes after IR. Ratios of PAR signal intensities relative to the GAPDH signals, normalized to the ratio in untreated (Ctrl) cells. The data represent the mean ± SEM, n = 7. *, P ≤ 0.05 (paired t test). C, REH cells (1 × 106 cells/mL) were treated with or without olaparib (10 μmol/L) for 30 minutes, followed by treatment with forsk and IR as described in A. Cells were harvested 5 minutes after IR, and lysates were subjected to immunoblot analysis with antibodies against PAR and GAPDH. Left, One representative Western blot analysis of three is shown. Right, Ratio of PAR signal intensities relative to the GAPDH signals, normalized to the ratio in the sample not treated with olaparib. The data represent the mean ± SEM, n = 3. *, P < 0.0001 (paired t test). D, REH cells (1 × 106 cells/mL) were incubated in the presence or absence of forsk (60 μmol/L) and doxorubicin (Dx, 200 nmol/L), and the cells were harvested 5 minutes after treatment with Dx. Total lysates were subjected to immunoblot analyses with antibodies against PAR and GAPDH. Left, One representative Western blot analysis of five is shown. Right, Ratios of PAR signal intensities relative to GAPDH signals, normalized to the ratio in untreated (Ctrl) cells. The data represent the mean ± SEM, n = 4. *, P < 0.05; **, P < 0.01 (paired t test). E, REH cells were treated in the presence or absence of 8-CPT-cAMP (200 μmol/L). Cells were harvested after 5 minutes and total lysates were subjected to immunoblot analyses with antibodies against PAR and GAPDH. Left, One representative Western blot analysis of four is shown. Right, Ratios of PAR signal intensities relative to GAPDH signals, normalized to the ratio in the sample not treated with 8-CPT-cAMP. The data represent the mean ± SEM, n = 4. *, P < 0.05 (paired t test).

Figure 3.

cAMP signaling and IR induces parylation in REH cells. A, REH cells (1 × 106 cells/mL) were incubated in the presence or absence of forskolin (forsk, 60 μmol/L) and IR (10 Gy), and the cells were harvested at the indicated timepoints. Total lysates were subjected to immunoblot analyses with antibodies against PAR and GAPDH. One representative Western blot analysis of seven is shown. B, The cells were treated as described in A and harvested 5 minutes after IR. Ratios of PAR signal intensities relative to the GAPDH signals, normalized to the ratio in untreated (Ctrl) cells. The data represent the mean ± SEM, n = 7. *, P ≤ 0.05 (paired t test). C, REH cells (1 × 106 cells/mL) were treated with or without olaparib (10 μmol/L) for 30 minutes, followed by treatment with forsk and IR as described in A. Cells were harvested 5 minutes after IR, and lysates were subjected to immunoblot analysis with antibodies against PAR and GAPDH. Left, One representative Western blot analysis of three is shown. Right, Ratio of PAR signal intensities relative to the GAPDH signals, normalized to the ratio in the sample not treated with olaparib. The data represent the mean ± SEM, n = 3. *, P < 0.0001 (paired t test). D, REH cells (1 × 106 cells/mL) were incubated in the presence or absence of forsk (60 μmol/L) and doxorubicin (Dx, 200 nmol/L), and the cells were harvested 5 minutes after treatment with Dx. Total lysates were subjected to immunoblot analyses with antibodies against PAR and GAPDH. Left, One representative Western blot analysis of five is shown. Right, Ratios of PAR signal intensities relative to GAPDH signals, normalized to the ratio in untreated (Ctrl) cells. The data represent the mean ± SEM, n = 4. *, P < 0.05; **, P < 0.01 (paired t test). E, REH cells were treated in the presence or absence of 8-CPT-cAMP (200 μmol/L). Cells were harvested after 5 minutes and total lysates were subjected to immunoblot analyses with antibodies against PAR and GAPDH. Left, One representative Western blot analysis of four is shown. Right, Ratios of PAR signal intensities relative to GAPDH signals, normalized to the ratio in the sample not treated with 8-CPT-cAMP. The data represent the mean ± SEM, n = 4. *, P < 0.05 (paired t test).

Close modal

Next, we explored whether the enhancing effects of cAMP signaling on IR-induced parylation also applied to other DNA-damaging scenarios. To this end, we used the same settings as presented in Fig. 3A—only replacing IR with the clinically relevant chemotherapeutic drug doxorubicin. As shown in Fig. 3D, the combination of forskolin and doxorubicin resulted in a 5-fold enhancement of parylation relative to the control.

Finally, we increased intracellular levels of cAMP by use of the cAMP-analog 8-CPT-cAMP to ensure that the effect of forskolin on parylation was not via cAMP-independent mechanisms. As shown in Fig. 3E, 8-CPT-cAMP enhanced parylation in the same manner as did forskolin.

cAMP signaling enhances DNA damage–induced levels of ROS

To reveal mechanisms involved in cAMP-mediated activation of PARP1, we explored the role of ROS. Intracellular ROS can be formed as a result of metabolic activity and play important roles in cell signaling and homeostasis maintenance (32). However, ROS can also arise from exogenous sources such as ionizing radiation (33). ROS has been implicated in regulation of autophagy (34), apoptosis (35), and PARP1 activation (36, 37). In the present experiments, ROS levels were assessed by flow cytometry after staining REH cells with the CellROX green reagent. We showed that forskolin or IR treatment alone increased the levels of ROS, and that forskolin enhanced the IR-induced ROS levels. Whereas the effects of forskolin were detectable after 30 minutes—with a peak after 2 hours, the IR-mediated elevation of ROS levels was somewhat slower (Fig. 4A). Furthermore, we demonstrated that the ROS scavenger NAC reduced the IR- and forskolin-induced intracellular levels of ROS (Fig. 4B). Notably, the same regulation of ROS was observed in primary leukemic cells from a patient with ALL (Fig. 4C).

Figure 4.

cAMP signaling increases ROS levels and depletes NAD in ALL cells. A, REH cells (4 × 105 cells/mL) were treated with or without forskolin (forsk, 60 μmol/L) for 45 minutes prior to IR (10 Gy). Cells were stained with CellROX green at the indicated timepoints. Data represent the mean ± SEM, n = 9. *, P < 0.05; **, P < 0.01 relative to the untreated (Ctrl) sample (paired t test). B, REH cells (4 × 105 cells/mL) were treated with or without NAC (30 mmol/L) for 30 minutes, followed by treatment with forsk (60 μmol/L) for 45 minutes prior to IR (10 Gy). The cells were stained with CellROX green 2 hours after IR. The data represent the mean ± SEM, n = 6. *, P < 0.05; **, P < 0.01 relative to the untreated (Ctrl) sample, §, P < 0.05; §§, P < 0.01 relative to the sample treated without NAC (paired t test). C, Leukemic cells (5 × 105 cells) from 1 patient diagnosed with ALL (ALL #65) were treated with or without NAC (30 mmol/L) for 30 minutes, followed by treatment with or without forsk (60 μmol/L) for 45 minutes prior to IR (5 Gy). The cells were harvested 2 hours after IR and stained with CellROX green. Data represent the MFI of one experiment. D, REH cells (1 × 106 cells/mL) were treated with or without NAC (30 mmol/L) for 30 minutes, followed by treatment with forsk (60 μmol/L) and IR (10 Gy). Cells were harvested 5 minutes after IR, and lysates were subjected to immunoblot analysis with antibodies against PAR and GAPDH. Left, One representative Western blot analysis of five is shown. Right, Ratio of PAR signal intensities relative to GAPDH signals, normalized to the ratio in the sample not treated with NAC. The data represent the mean ± SEM, n = 5. *, P = 0.0001 (paired t test). E, REH cells (4 × 105 cells/mL) were treated with or without NAC (30 mmol/L) for 30 minutes, followed by treatment with forsk (60 μmol/L) for 45 minutes prior to IR (10 Gy). The cells were stained with CYTO-ID 6 hours after IR. The data represent the mean ± SEM, n = 12. *, P < 0.05; **, P < 0.01 relative to untreated (Ctrl) sample, §, P < 0.05; §§, P < 0.01 relative to the sample treated without NAC (paired t test). F, Leukemic cells (5 × 105 cells) from 1 patient diagnosed with ALL (ALL #75) were treated with or without NAC (30 mmol/L) for 30 minutes, followed by treatment with or without forsk (60 μmol/L) for 45 minutes prior to IR (5 Gy). The cells were stained with CYTO-ID 6 hours after IR. G, REH cells (4 × 105 cells/mL) were treated with NAC, forsk and IR as described in E. The cells were stained with PI 24 hours after IR. The data represent the mean ± SEM, n = 6. *, P < 0.05; **, P < 0.01 relative to the untreated (Ctrl) sample, §, P < 0.05; §§, P < 0.01 relative to the sample treated without NAC (paired t test). H, REH cells (2 × 106 cells) were treated with or without olaparib (40 μmol/L) or NAC (30 mmol/L) for 30 minutes, followed by treatment with or without forskolin (forsk, 60 μmol/L) and IR (10 Gy; left) or doxorubicin (400 nmol/L; right) for 30 minutes. The cells were harvested, and NAD levels were quantified as described in “Supplementary Materials and Methods.” Data represent the mean ± SEM, n = 4. *, P < 0.05; **, P < 0.01 (paired t test). I and J, REH cells (4 × 105 cells/mL) were treated with or without NAD at the indicated concentrations for 30 minutes, followed by treatment with forsk (60 μmol/L) for 45 minutes prior to IR (10 Gy). I, The cells were stained with CYTO-ID 6 hours after IR. The data represent the mean ± SEM, n = 5. *, P < 0.05; **, P < 0.01 relative to untreated (Ctrl) sample, §, P < 0.05; §§, P < 0.01 relative to the sample treated without NAD (paired t test). J, Cells were stained with PI 24 hours after IR. The data represent the mean ± SEM, n = 5. *, P < 0.05 (paired t test).

Figure 4.

cAMP signaling increases ROS levels and depletes NAD in ALL cells. A, REH cells (4 × 105 cells/mL) were treated with or without forskolin (forsk, 60 μmol/L) for 45 minutes prior to IR (10 Gy). Cells were stained with CellROX green at the indicated timepoints. Data represent the mean ± SEM, n = 9. *, P < 0.05; **, P < 0.01 relative to the untreated (Ctrl) sample (paired t test). B, REH cells (4 × 105 cells/mL) were treated with or without NAC (30 mmol/L) for 30 minutes, followed by treatment with forsk (60 μmol/L) for 45 minutes prior to IR (10 Gy). The cells were stained with CellROX green 2 hours after IR. The data represent the mean ± SEM, n = 6. *, P < 0.05; **, P < 0.01 relative to the untreated (Ctrl) sample, §, P < 0.05; §§, P < 0.01 relative to the sample treated without NAC (paired t test). C, Leukemic cells (5 × 105 cells) from 1 patient diagnosed with ALL (ALL #65) were treated with or without NAC (30 mmol/L) for 30 minutes, followed by treatment with or without forsk (60 μmol/L) for 45 minutes prior to IR (5 Gy). The cells were harvested 2 hours after IR and stained with CellROX green. Data represent the MFI of one experiment. D, REH cells (1 × 106 cells/mL) were treated with or without NAC (30 mmol/L) for 30 minutes, followed by treatment with forsk (60 μmol/L) and IR (10 Gy). Cells were harvested 5 minutes after IR, and lysates were subjected to immunoblot analysis with antibodies against PAR and GAPDH. Left, One representative Western blot analysis of five is shown. Right, Ratio of PAR signal intensities relative to GAPDH signals, normalized to the ratio in the sample not treated with NAC. The data represent the mean ± SEM, n = 5. *, P = 0.0001 (paired t test). E, REH cells (4 × 105 cells/mL) were treated with or without NAC (30 mmol/L) for 30 minutes, followed by treatment with forsk (60 μmol/L) for 45 minutes prior to IR (10 Gy). The cells were stained with CYTO-ID 6 hours after IR. The data represent the mean ± SEM, n = 12. *, P < 0.05; **, P < 0.01 relative to untreated (Ctrl) sample, §, P < 0.05; §§, P < 0.01 relative to the sample treated without NAC (paired t test). F, Leukemic cells (5 × 105 cells) from 1 patient diagnosed with ALL (ALL #75) were treated with or without NAC (30 mmol/L) for 30 minutes, followed by treatment with or without forsk (60 μmol/L) for 45 minutes prior to IR (5 Gy). The cells were stained with CYTO-ID 6 hours after IR. G, REH cells (4 × 105 cells/mL) were treated with NAC, forsk and IR as described in E. The cells were stained with PI 24 hours after IR. The data represent the mean ± SEM, n = 6. *, P < 0.05; **, P < 0.01 relative to the untreated (Ctrl) sample, §, P < 0.05; §§, P < 0.01 relative to the sample treated without NAC (paired t test). H, REH cells (2 × 106 cells) were treated with or without olaparib (40 μmol/L) or NAC (30 mmol/L) for 30 minutes, followed by treatment with or without forskolin (forsk, 60 μmol/L) and IR (10 Gy; left) or doxorubicin (400 nmol/L; right) for 30 minutes. The cells were harvested, and NAD levels were quantified as described in “Supplementary Materials and Methods.” Data represent the mean ± SEM, n = 4. *, P < 0.05; **, P < 0.01 (paired t test). I and J, REH cells (4 × 105 cells/mL) were treated with or without NAD at the indicated concentrations for 30 minutes, followed by treatment with forsk (60 μmol/L) for 45 minutes prior to IR (10 Gy). I, The cells were stained with CYTO-ID 6 hours after IR. The data represent the mean ± SEM, n = 5. *, P < 0.05; **, P < 0.01 relative to untreated (Ctrl) sample, §, P < 0.05; §§, P < 0.01 relative to the sample treated without NAD (paired t test). J, Cells were stained with PI 24 hours after IR. The data represent the mean ± SEM, n = 5. *, P < 0.05 (paired t test).

Close modal

ROS is involved in cAMP-mediated PARP1 parylation, autophagy, and survival of ALL cells

To address the role of ROS in cAMP-mediated parylation of PARP1, REH cells were pretreated with NAC for 30 minutes prior to IR and forskolin exposure. NAC reduced the parylation of PARP1 by 77% relative to untreated cells, which indicates that ROS are indeed involved in PARP1 parylation in REH cells (Fig. 4D).

With ROS being involved in cAMP-mediated parylation of PARP1, we anticipated that ROS would have the same role as PARP1 in cAMP-mediated autophagy and inhibition of DNA damage–induced cell death. To address this issue, we again utilized NAC to reduce ROS levels. We demonstrated that NAC significantly reduced the autophagy induced by both forskolin and IR in REH cells (Fig. 4E) as well as in primary cells derived from a patient with ALL (Fig. 4F). Importantly, we revealed that NAC enhanced the IR-mediated killing of the cells protected by forskolin, and thereby reversed the forskolin-mediated inhibition of DNA damage–induced cell death by approximately 50% (Fig. 4G). Taken together, these results suggest that ROS has a pivotal role in the cAMP-mediated induction of autophagy as well as in the subsequent survival of the cells.

In an attempt to explain how cAMP signaling enhances the levels of ROS in ALL cells, we addressed the role of NADPH oxidase (NOX)—an enzyme family known to generate intracellular ROS by transferring one electron from NADPH to oxygen (38). By taking advantage of the established NOX inhibitor VAS2870, we found that neither the cAMP-mediated induction of autophagy nor the inhibition of DNA damage–induced cell death involved the activation of NOX (Supplementary Fig. S2).

cAMP-mediated autophagy and survival of ALL cells involves reduction in NAD

NAD is one of the substrates required for cells to execute parylation, and increased parylation will consequently deprive cells of NAD causing energy deprivation and starvation (39, 40). We therefore hypothesized that ROS-mediated parylation induced by cAMP signaling and/or IR/doxorubicin would result in reduced levels of NAD. Furthermore, we assumed that inhibiting parylation or reducing ROS would restore NAD levels. In agreement with our hypothesis, we found that cotreatment of the cells with forskolin and IR (Fig. 4H, left) or forskolin and doxorubicin (Fig. 4H, right) significantly reduced the levels of NAD compared with untreated control cells, and that addition of olaparib or NAC reversed these effects.

As a putative starvation signal, it is acknowledged that reduced levels of NAD may induce autophagy (41–43). Accordingly, we tested whether maintaining elevated cellular levels of NAD would counter the effects of forskolin on DNA damage–induced autophagy and cell death. Autophagy induced by cotreating the cells with forskolin and IR was inhibited by 38% and 48% in the presence of 10 and 20 mmol/L NAD, respectively (Fig. 4I). Furthermore, the same concentrations of NAD reversed the cAMP-mediated protection against IR-induced cell death from 38% to −14% (10 mmol/L NAD) and −34% (20 mmol/L NAD; Fig. 4J).

PARP1 inhibition enhances the effect of DNA-damaging treatment on ALL progression in vivo

We previously proposed that targeting cAMP-mediated autophagy would abolish bone marrow–mediated protection against DNA damage–induced cell death and thereby be useful in treatment of ALL (4–8, 14). Our current findings demonstrate a critical role of PARP1 in these processes, suggesting that targeting PARP1 could be a novel strategy to improve the effect of conventional genotoxic treatment of ALL. This possibility was tested in our recently established xenograft model of ALL in NSG mice (14). REH cells were lentivirally transduced to express a vector containing firefly luciferase and EGFP, before the cells were subjected to intratibial injection into the NSG mice. After successful REH cell engraftment, the animals were treated with whole-body IR in the presence or absence of olaparib, Fig. 5A shows a timeline of the various treatments and the in vivo imaging procedures performed in the xenograft experiment presented in Fig. 5B and C. The experiments were terminated approximately 30 days after initiation, due to the mice showing intolerable signs of illness. As shown in the left panel of Fig. 5B, olaparib over time potentiated the inhibiting effects of IR on ALL progression. Moreover, IR-treated animals in the absence of olaparib revealed more disseminated disease (Fig. 5B, right panel indicated by red arrows) than the animals cotreated with olaparib. Quantitatively, olaparib reduced the luminescence from the leukemic cells in IR-treated mice by 34% as compared with animals treated with IR alone (Fig. 5B and C). Finally, we also established a xenograft model of primary leukemic cells derived from a patient diagnosed with ALL. Animals treated with olaparib or vehicle only, showed early intolerable signs of illness and had to be euthanized. Accordingly, we could only follow the progression of leukemia over time in whole-body irradiated mice treated with or without olaparib. In vivo imaging of these mice was performed weekly from 15 weeks post intratibial injection. As shown in Fig. 5D, the average luminescence signal was reduced by 63% at the end of the experiment when the mice were cotreated with olaparib compared with IR alone (Fig. 5D). The beneficial effect of olaparib on leukemia progression was confirmed by measuring the levels of CD19+CD10+ leukemic cells in blood samples taken from the mice at the end point of the experiment. Olaparib reduced the leukemia burden in IR-treated mice from 17% to 5% (Fig. 5E).

Figure 5.

Olaparib potentiates the effect of IR on ALL cancer progression in vivo. A, Timeline for the in vivo experiment presented in Fig. 6B and C: Intratibial (i.t) injections of transduced ALL cells were performed at day 0. In vivo imaging, IR, and injections of vehicle (10% 2-hydroxpropyl-beta-cyclodextrin, HBC in PBS) or olaparib (50 mg/kg in 10% HBC in PBS) commenced at day 5. In vivo imaging was performed twice weekly, vehicle/olaparib injections thrice weekly, and IR once per week. Treatments lasted for 3 weeks. B, Xenograft REH mice (5 mice per treatment group) were treated with IR in the presence or absence of olaparib as described in Fig. 6A and “Supplementary Materials and Methods.” Development of leukemia was followed by noninvasive in vivo imaging of luminescence emitted from the mice. Left, Luminescence image of xenograft REH mice taken at the experimental endpoint (30 days after intratibial injection). Tumor disseminations are indicated by red arrows. Right, Xenograft luciferase activity [photons per second, (p/s)] over time. Each data point represents the mean ± SEM signal intensity of the 5 xenograft mice in each treatment group. C, Luminescence intensity at the experimental endpoint of three equal experiments of xenograft REH mice treated with IR and with or without olaparib as described in Fig. 6A. Each symbol represents the luminescence intensity of 1 mouse in one experiment. The different experiments are indicated by different symbols (circle, triangle, rectangle), n = 15. *, P < 0.05 (unpaired t test). D and E, Xenograft ALL#71 mice (5 mice per treatment group) were treated with IR and with or without olaparib as described in Fig. 6A. D, Development of leukemia was followed by noninvasive in vivo imaging of luminescence from the mice. Left, Luminescence image of xenograft ALL#71 mice taken at the experimental endpoint (17 weeks after intratibial injection). Right, Xenograft luciferase activities (p/s) were measured at the indicated timepoints after intratibial injection. The values are relative to the flux intensity recorded 15 weeks post intratibialinjection. Each symbol represents the average luminescence intensity from 5 mice. The vertical lines represent the SEM. P < 0.05 (paired t test). E, Bone marrow cells were harvested from xenograft ALL#71 mice at the end of the experiment, stained with fluorochrome-conjugated CD19 and CD10 antibodies, and then analyzed by flow cytometry. Left, Representative flow cytometry dot plots for bone marrow cells isolated from 1 mouse treated with IR and 1 mouse treated with IR and olaparib. Right, Each symbol represents the percentage of CD19+CD10+ cells present in the bone marrow of one animal, n = 5. The horizontal lines represent the mean. *, P < 0.01 (paired t test).

Figure 5.

Olaparib potentiates the effect of IR on ALL cancer progression in vivo. A, Timeline for the in vivo experiment presented in Fig. 6B and C: Intratibial (i.t) injections of transduced ALL cells were performed at day 0. In vivo imaging, IR, and injections of vehicle (10% 2-hydroxpropyl-beta-cyclodextrin, HBC in PBS) or olaparib (50 mg/kg in 10% HBC in PBS) commenced at day 5. In vivo imaging was performed twice weekly, vehicle/olaparib injections thrice weekly, and IR once per week. Treatments lasted for 3 weeks. B, Xenograft REH mice (5 mice per treatment group) were treated with IR in the presence or absence of olaparib as described in Fig. 6A and “Supplementary Materials and Methods.” Development of leukemia was followed by noninvasive in vivo imaging of luminescence emitted from the mice. Left, Luminescence image of xenograft REH mice taken at the experimental endpoint (30 days after intratibial injection). Tumor disseminations are indicated by red arrows. Right, Xenograft luciferase activity [photons per second, (p/s)] over time. Each data point represents the mean ± SEM signal intensity of the 5 xenograft mice in each treatment group. C, Luminescence intensity at the experimental endpoint of three equal experiments of xenograft REH mice treated with IR and with or without olaparib as described in Fig. 6A. Each symbol represents the luminescence intensity of 1 mouse in one experiment. The different experiments are indicated by different symbols (circle, triangle, rectangle), n = 15. *, P < 0.05 (unpaired t test). D and E, Xenograft ALL#71 mice (5 mice per treatment group) were treated with IR and with or without olaparib as described in Fig. 6A. D, Development of leukemia was followed by noninvasive in vivo imaging of luminescence from the mice. Left, Luminescence image of xenograft ALL#71 mice taken at the experimental endpoint (17 weeks after intratibial injection). Right, Xenograft luciferase activities (p/s) were measured at the indicated timepoints after intratibial injection. The values are relative to the flux intensity recorded 15 weeks post intratibialinjection. Each symbol represents the average luminescence intensity from 5 mice. The vertical lines represent the SEM. P < 0.05 (paired t test). E, Bone marrow cells were harvested from xenograft ALL#71 mice at the end of the experiment, stained with fluorochrome-conjugated CD19 and CD10 antibodies, and then analyzed by flow cytometry. Left, Representative flow cytometry dot plots for bone marrow cells isolated from 1 mouse treated with IR and 1 mouse treated with IR and olaparib. Right, Each symbol represents the percentage of CD19+CD10+ cells present in the bone marrow of one animal, n = 5. The horizontal lines represent the mean. *, P < 0.01 (paired t test).

Close modal

Over the past few decades, survival of children diagnosed with ALL has significantly improved. However, treatment resistance and severe long-term side effects of current multimodal chemotherapy are still major problems (2, 3). In the attempt to identify targets that may improve the effect of current treatment of ALL, we have in the current study investigated the mechanisms whereby cAMP signaling enhances autophagy and thereby inhibits DNA damage–induced killing of ALL cells. To induce DNA damage in the cells, we have primarily exposed the cells to ionizing IR. In a clinical setting of treating ALL, IR is generally limited to patients with dissemination of the disease to the central nervous system or the testicles, but we have previously shown that cAMP signaling exerts similar inhibitory effects on DNA damage responses independently of whether the DNA damage is inflicted by IR or chemotherapeutic drugs commonly used to treat ALL (4). However, to expand the clinical relevance of our results, we also included the chemotherapeutic agent doxorubicin. IR and doxorubicin both induce DNA damage via formation of ROS (44, 45). Although ROS-induced DNA damage is a common mechanism for many cytotoxic cancer drugs (46), we cannot exclude the possibility that ROS-induced autophagy might be less imperative for cell death induced by other types of DNA-damaging agents. Nevertheless, we reveal that cAMP-mediated autophagy is preceded by ROS-induced activation of PARP1, and that the PARP1 inhibitor olaparib enhances the DNA damage–mediated killing of ALL cells by abolishing this pathway. With this, we put forward PARP1 as a novel target to increase the therapeutic index of current DNA-damaging chemotherapy regimens in treatment of ALL.

Given the paramount role of PARP1 in DNA damage responses (15, 20, 21, 37), it was not surprising to find that DNA damage induced by either IR or doxorubicin resulted in rapid activation of PARP1 in the ALL cells. More interestingly however, was the finding that elevation of intracellular levels of cAMP by forskolin or by the cAMP analog 8-CPT-cAMP induced activation of PARP1, and that forskolin further enhanced the DNA damage–induced PARP1 activation. Activation of PARP1 has been associated with induction of autophagy (19–21, 37), and we found that both the induced autophagy and the enhanced survival promoted by cAMP signaling involved activation of PARP1. An interesting publication by Rodríguez-Vargas and coworkers showed that autophagy in MCF7 cells required PARP1-mediated parylation and activation of the autophagy-initiating enzyme AMP-activated protein kinase (AMPK; ref. 21). We demonstrated that forskolin induced a rapid and transient activation of AMPK (Supplementary Fig. S3A), as well as of its substrate unc like kinase 1 (ULK1) in REH cells (Supplementary Fig. S3B). However, we did not observe any parylation of AMPK (Supplementary Fig. S3C). Instead, we revealed the importance of NAD in the present parylation process, by demonstrating that PARP1 inhibition prevented IR- and forskolin-mediated depletion of NAD. The connection between PARP1 inhibition and NAD depletion was also recently suggested by Somers and coworkers, by demonstrating that inhibition of nicotinamide phosphoribosyl transferase in ALL cells indirectly inhibits PARP1 via depletion of NAD (47). Here we showed that exogenously added NAD reduced autophagy induced by the combination of IR and forskolin treatment. Moreover, we demonstrated that exogenously added NAD to the cell cultures impaired the ability of forskolin to inhibit DNA damage–induced killing of ALL cells. Together, these data suggest that enhanced cAMP-mediated protein parylation deprives the cells of NAD, which in turn results in enhanced autophagy and cell survival.

To explain how cAMP-signaling activates PARP1 in ALL cells, we addressed the role of ROS. This approach was chosen, because ROS are among the established activators of PARP1 (36, 37), and because enhanced levels of ROS are reported both upstream and downstream of DNA-damaging lesions (48). As expected, both IR and doxorubicin enhanced the levels of ROS. Notably, forskolin alone also increased intracellular ROS levels, and further enhanced the levels of IR-induced ROS both in REH cells and in primary leukemic cells from a patient with ALL. The ROS scavenger NAC reduced the parylation of PARP1 in REH cells treated with IR and forskolin. NAC also reduced the cAMP-induced autophagy, and it counteracted the inhibitory effect of forskolin on DNA damage–induced killing of ALL cells. Thus, our results support the notion that ROS is required for cAMP-mediated activation of PARP1, which in turn results in enhanced autophagy and thereby to increased survival of leukemic cells. In support of our results, it was shown that activation of mitochondrial PARP1 is involved in oxidant-induced killing of the monocytic cell line U937 (49). Interestingly, activation of the cAMP signaling pathway enhanced the oxidant-mediated PARP1 activation in the U937 cells, whereas inhibition of the pathway protected against the oxidant-induced cell injury (49). Although our current results support the concept that cAMP-induced ROS prevent DNA damage–induced apoptosis of ALL cells via PARP1-mediated autophagy, we cannot exclude the possibility that ROS could exert its effect also via other mechanisms. Previously we showed that cAMP-mediated inhibition of IR-induced apoptosis in ALL cells involves the down regulation of the key protein p53 (4, 5, 7). In the current study, we demonstrated that neither treatment with NAC nor olaparib affected the levels of p53 (Supplementary Fig. S4), supporting our recent finding that cAMP-mediated enhancement of DNA damage–induced autophagy is a p53-independent process (8).

To reveal the mechanisms whereby cAMP signaling enhances the levels of ROS in ALL cells, we used the established NOX2) inhibitor VAS2870 to rule out the possible involvement of NOX2-induced ROS in cAMP-mediated autophagy and enhanced survival of DNA-damaged cells. The role of the antioxidant NAC as a ROS scavenger, is generally explained by its ability to act as a precursor for the naturally occurring antioxidant glutathione (50). In light of the selective ability of NAC and not VAS2870 to counter the cAMP-mediated processes in the ALL cells, it might be worthwhile to investigate whether suppression of antioxidant systems such as superoxide dismutase, catalase, or glutathione could be involved in the cAMP-mediated induction of ROS. Alternatively, the enhanced ROS levels could be the result of increased metabolic activity (32, 51). However, we consider the latter possibility as an unlikely mechanism, because we have previously established that cAMP signaling promotes cell-cycle arrest and results in reduced proliferation of ALL cells (52–54).

PARP1 inhibitors are already incorporated in treatment of patients suffering from cancers such as ovarian and metastatic breast cancer (55). These cancers are characterized by deficiency in their double-strand break repair system through mutations in the BRCA1 and BRCA2 genes. BRCA mutations have not been linked to ALL, but it is interesting to note that ALL cells frequently possess deficiencies in their DNA damage-repair systems (56). It has been shown that combining PARP1 inhibition with IR or other DNA-damaging agents, sensitizes BRCA-mutated cancer cells to cell death (57–59), and we cannot rule out the possibility of synthetic lethality imposed by PARP1 inhibitors also in ALL cells. However, in the current study, we have demonstrated that inhibitors of PARP1 enhances DNA damage–induced killing of ALL cells by opposing the ability of cAMP signaling to enhance autophagy. Figure 6 summarizes the mechanisms on how cAMP signaling enhances DNA damage–induced autophagy and thereby reduces the killing of ALL cells, and it also explains how PARP1 inhibitors may interfere with this pathway. According to our model, cAMP signaling enhances ROS levels induced by DNA-damaging treatment, which in turn leads to PARP1 activation and parylation of target proteins. The resulting depletion of endogenous levels of NAD promotes the induction of autophagy, which has a protective role against DNA damage–induced apoptosis. Concurrent with our model, the accumulation of ROS is inhibited by NAC, whereas olaparib inhibits activation of PARP1 by preventing the depletion of NAD.

Figure 6.

A model explaining the involvement of ROS and PARP1 in cAMP-mediated autophagy and prevention of DNA damage–induced killing of ALL cells. cAMP signaling enhances the DNA damage–induced intracellular levels of ROS. This results in activation of PARP1, which in turn leads to parylation of target proteins and thus to depletion of NAD—one of the substrates required for protein parylation. The depletion of NAD causes downstream activation of autophagy and thereby inhibition of DNA damage–induced apoptosis in a p53-independent manner. cAMP-mediated autophagy can be reversed through inhibition of ROS via NAC, or it can be countered by restoring the NAD levels through the PARP1 inhibitor olaparib, or by exogenous addition of NAD.

Figure 6.

A model explaining the involvement of ROS and PARP1 in cAMP-mediated autophagy and prevention of DNA damage–induced killing of ALL cells. cAMP signaling enhances the DNA damage–induced intracellular levels of ROS. This results in activation of PARP1, which in turn leads to parylation of target proteins and thus to depletion of NAD—one of the substrates required for protein parylation. The depletion of NAD causes downstream activation of autophagy and thereby inhibition of DNA damage–induced apoptosis in a p53-independent manner. cAMP-mediated autophagy can be reversed through inhibition of ROS via NAC, or it can be countered by restoring the NAD levels through the PARP1 inhibitor olaparib, or by exogenous addition of NAD.

Close modal

To assess the therapeutic potential of PARP1 inhibitors in combination with DNA-damaging treatment of ALL, we expanded our previously established ALL xenograft model by irradiating the mice in the presence and absence of olaparib. To reveal putative effects of olaparib, we used suboptimal IR doses that only slowly reduced leukemia progression. It should be noted that olaparib alone in certain mice slightly promoted the progression of ALL (Fig. 5). However, in the presence of IR, olaparib consistently enhanced the effects of the suboptimal doses of IR and thereby reduced the leukemia progression—both in the REH xenograft mice and in the mice engrafted with patient-derived primary ALL cells. Hence, it is tempting to propose that PARP inhibitors may be beneficial in a clinical setting of ALL to increase the sensitivity of DNA-damaging therapy to reduce the dosages, and thereby limit the severe side-effects of current treatments.

H.K. Blomhoff reports grants from University of Oslo, UNIFOR Frimed Foundation, The Norwegian Cancer Society, The Norwegian Childhood Cancer Foundation, The Jahre Foundation, The Blix Family Foundation, and The Nansen Foundation during the conduct of the study. No disclosures were reported by the other authors.

N. Richartz: Data curation, software, validation, investigation, visualization, methodology, writing–original draft, writing–review and editing. W. Pietka: Formal analysis, validation, investigation, methodology, writing–review and editing. K.M. Gilljam: Formal analysis, validation, investigation, methodology, writing–review and editing. S. Skah: Validation, investigation, writing–review and editing. B.S. Skalhegg: Data curation, formal analysis, writing–review and editing. S. Bhagwat: Investigation, writing–review and editing. E.H. Naderi: Supervision, validation, investigation, writing–review and editing. E. Ruud: Resources, supervision, writing–review and editing. H.K. Blomhoff: Conceptualization, resources, supervision, funding acquisition, validation, methodology, writing–original draft, project administration, writing–review and editing.

We thank Pietri Puustinen and Jonas Aakre Wik for excellent technical assistance and fruitful discussions. This work was supported by grants from the Norwegian Cancer Society, The Norwegian Childhood Cancer Society, The Anders Jahre Foundation, The Blix Family Foundation, The Nansen Foundation, the UNIFOR FRIMED Foundation, and the University of Oslo.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Greaves
M
.
A causal mechanism for childhood acute lymphoblastic leukaemia
.
Nat Rev Cancer
2018
;
18
:
471
84
.
2.
Curtin
SC
,
Minino
AM
,
Anderson
RN
.
Declines in cancer death rates among children and adolescents in the United States, 1999–2014
.
NCHS Data Brief
2016
:
1
8
.
3.
Ness
KK
,
Armenian
SH
,
Kadan-Lottick
N
,
Gurney
JG
.
Adverse effects of treatment in childhood acute lymphoblastic leukemia: general overview and implications for long-term cardiac health
.
Expert Rev Hematol
2011
;
4
:
185
97
.
4.
Naderi
EH
,
Findley
HW
,
Ruud
E
,
Blomhoff
HK
,
Naderi
S
.
Activation of cAMP signaling inhibits DNA damage-induced apoptosis in BCP-ALL cells through abrogation of p53 accumulation
.
Blood
2009
;
114
:
608
18
.
5.
Naderi
EH
,
Jochemsen
AG
,
Blomhoff
HK
,
Naderi
S
.
Activation of cAMP signaling interferes with stress-induced p53 accumulation in ALL-derived cells by promoting the interaction between p53 and HDM2
.
Neoplasia
2011
;
13
:
653
63
.
6.
Naderi
EH
,
Skah
S
,
Ugland
H
,
Myklebost
O
,
Sandnes
DL
,
Torgersen
ML
, et al
.
Bone marrow stroma-derived PGE2 protects BCP-ALL cells from DNA damage-induced p53 accumulation and cell death
.
Mol Cancer
2015
;
14
:
14
.
7.
Naderi
EH
,
Ugland
HK
,
Diep
PP
,
Josefsen
D
,
Ruud
E
,
Naderi
S
, et al
.
Selective inhibition of cell death in malignant vs normal B-cell precursors: implications for cAMP in development and treatment of BCP-ALL
.
Blood
2013
;
121
:
1805
13
.
8.
Skah
S
,
Richartz
N
,
Duthil
E
,
Gilljam
KM
,
Bindesboll
C
,
Naderi
EH
, et al
.
cAMP-mediated autophagy inhibits DNA damage-induced death of leukemia cells independent of p53
.
Oncotarget
2018
;
9
:
30434
49
.
9.
Yin
T
,
Li
L
.
The stem cell niches in bone
.
J Clin Invest
2006
;
116
:
1195
201
.
10.
Walkley
CR
,
Shea
JM
,
Sims
NA
,
Purton
LE
,
Orkin
SH
.
Rb regulates interactions between hematopoietic stem cells and their bone marrow microenvironment
.
Cell
2007
;
129
:
1081
95
.
11.
McMillin
DW
,
Delmore
J
,
Weisberg
E
,
Negri
JM
,
Geer
DC
,
Klippel
S
, et al
.
Tumor cell-specific bioluminescence platform to identify stroma-induced changes to anticancer drug activity
.
Nat Med
2010
;
16
:
483
9
.
12.
Denizot
Y
,
Donnard
M
,
Truffinet
V
,
Malissein
E
,
Faucher
JL
,
Turlure
P
, et al
.
Functional EP2 receptors on blast cells of patients with acute leukemia
.
Int J Cancer
2005
;
115
:
499
501
.
13.
Malissein
E
,
Reynaud
S
,
Bordessoule
D
,
Faucher
JL
,
Turlure
P
,
Trimoreau
F
, et al
.
PGE(2) receptor subtype functionality on immature forms of human leukemic blasts
.
Leuk Res
2006
;
30
:
1309
13
.
14.
Richartz
N
,
Duthil
E
,
Ford
A
,
Naderi
EH
,
Bhagwat
S
,
Gilljam
KM
, et al
.
Targeting cyclooxygenase by indomethacin decelerates progression of acute lymphoblastic leukemia in a xenograft model
.
Blood Adv
2019
;
3
:
3181
90
.
15.
Ray Chaudhuri
A
,
Nussenzweig
A
.
The multifaceted roles of PARP1 in DNA repair and chromatin remodelling
.
Nat Rev Mol Cell Biol
2017
;
18
:
610
21
.
16.
Tulin
A
,
Chinenov
Y
,
Spradling
A
.
Regulation of chromatin structure and gene activity by poly(ADP-ribose) polymerases
.
Curr Top Dev Biol
2003
;
56
:
55
83
.
17.
Oei
SL
,
Keil
C
,
Ziegler
M
.
Poly(ADP-ribosylation) and genomic stability
.
Biochem Cell Biol
2005
;
83
:
263
9
.
18.
Hottiger
MO
,
Boothby
M
,
Koch-Nolte
F
,
Lüscher
B
,
Martin
NMB
,
Plummer
R
, et al
.
Progress in the function and regulation of ADP-ribosylation
.
Sci Signal
2011
;
4
:
mr5
.
19.
Muñoz-Gámez
JA
,
Rodríguez-Vargas
JM
,
Quiles-Pérez
R
,
Aguilar-Quesada
R
,
Martín-Oliva
D
,
de Murcia
G
, et al
.
PARP-1 is involved in autophagy induced by DNA damage
.
Autophagy
2009
;
5
:
61
74
.
20.
Rodríguez-Vargas
JM
,
Oliver-Pozo
FJ
,
Dantzer
F
.
PARP1 and poly(ADP-ribosyl)ation signaling during autophagy in response to nutrient deprivation
.
Oxid Med Cell Longev
2019
;
2019
:
2641712
.
21.
Rodríguez-Vargas
JM
,
Rodríguez
MI
,
Majuelos-Melguizo
J
,
García-Diaz
Á
,
González-Flores
A
,
López-Rivas
A
, et al
.
Autophagy requires poly(adp-ribosyl)ation-dependent AMPK nuclear export
.
Cell Death Differ
2016
;
23
:
2007
18
.
22.
Rosenfeld
C
,
Goutner
A
,
Venuat
AM
,
Choquet
C
,
Pico
JL
,
Dore
JF
, et al
.
An effect human leukaemic cell line: Reh
.
Eur J Cancer
1977
;
13
:
377
9
.
23.
Klionsky
DJ
,
Abdelmohsen
K
,
Abe
A
,
Abedin
MJ
,
Abeliovich
H
,
Acevedo Arozena
A
, et al
.
Guidelines for the use and interpretation of assays for monitoring autophagy (3rd edition)
.
Autophagy
2016
;
12
:
1
222
.
24.
Guo
S
,
Liang
Y
,
Murphy
SF
,
Huang
A
,
Shen
H
,
Kelly
DF
, et al
.
A rapid and high content assay that measures cyto-ID-stained autophagic compartments and estimates autophagy flux with potential clinical applications
.
Autophagy
2015
;
11
:
560
72
.
25.
Perelman
A
,
Wachtel
C
,
Cohen
M
,
Haupt
S
,
Shapiro
H
,
Tzur
A
.
JC-1: alternative excitation wavelengths facilitate mitochondrial membrane potential cytometry
.
Cell Death Dis
2012
;
3
:
e430
.
26.
Jungmichel
S
,
Rosenthal
F
,
Altmeyer
M
,
Lukas
J
,
Hottiger
MO
,
Nielsen
ML
.
Proteome-wide identification of poly(ADP-Ribosyl)ation targets in different genotoxic stress responses
.
Mol Cell
2013
;
52
:
272
85
.
27.
Zhang
Y
,
Wang
J
,
Ding
M
,
Yu
Y
.
Site-specific characterization of the Asp- and Glu-ADP-ribosylated proteome
.
Nat Methods
2013
;
10
:
981
4
.
28.
Gibson
BA
,
Zhang
Y
,
Jiang
H
,
Hussey
KM
,
Shrimp
JH
,
Lin
H
, et al
.
Chemical genetic discovery of PARP targets reveals a role for PARP-1 in transcription elongation
.
Science
2016
;
353
:
45
50
.
29.
Alvarez-Gonzalez
R
,
Althaus
FR
.
Poly(ADP-ribose) catabolism in mammalian cells exposed to DNA-damaging agents
.
Mutat Res
1989
;
218
:
67
74
.
30.
Davidovic
L
,
Vodenicharov
M
,
Affar
EB
,
Poirier
GG
.
Importance of poly(ADP-ribose) glycohydrolase in the control of poly(ADP-ribose) metabolism
.
Exp Cell Res
2001
;
268
:
7
13
.
31.
Krüger
A
,
Bürkle
A
,
Hauser
K
,
Mangerich
A
.
Real-time monitoring of PARP1-dependent PARylation by ATR-FTIR spectroscopy
.
Nat Commun
2020
;
11
:
2174
.
32.
Nathan
C
,
Ding
A
.
SnapShot: reactive oxygen intermediates (ROI)
.
Cell
2010
;
140
:
951
951.e2
.
33.
Kohen
R
,
Nyska
A
.
Oxidation of biological systems: oxidative stress phenomena, antioxidants, redox reactions, and methods for their quantification
.
Toxicol Pathol
2002
;
30
:
620
50
.
34.
Scherz-Shouval
R
,
Shvets
E
,
Fass
E
,
Shorer
H
,
Gil
L
,
Elazar
Z
.
Reactive oxygen species are essential for autophagy and specifically regulate the activity of Atg4
.
EMBO J
2007
;
26
:
1749
60
.
35.
Perillo
B
,
Di Donato
M
,
Pezone
A
,
Di Zazzo
E
,
Giovannelli
P
,
Galasso
G
, et al
.
ROS in cancer therapy: the bright side of the moon
.
Exp Mol Med
2020
;
52
:
192
203
.
36.
Hocsak
E
,
Szabo
V
,
Kalman
N
,
Antus
C
,
Cseh
A
,
Sumegi
K
, et al
.
PARP inhibition protects mitochondria and reduces ROS production via PARP-1-ATF4-MKP-1-MAPK retrograde pathway
.
Free Radic Biol Med
2017
;
108
:
770
84
.
37.
Rodríguez-Vargas
JM
,
Ruiz-Magaña
MJ
,
Ruiz-Ruiz
C
,
Majuelos-Melguizo
J
,
Peralta-Leal
A
,
Rodríguez
MI
, et al
.
ROS-induced DNA damage and PARP-1 are required for optimal induction of starvation-induced autophagy
.
Cell Res
2012
;
22
:
1181
98
.
38.
Bedard
K
,
Krause
KH
.
The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology
.
Physiol Rev
2007
;
87
:
245
313
.
39.
Alemasova
EE
,
Lavrik
OI
.
Poly(ADP-ribosyl)ation by PARP1: reaction mechanism and regulatory proteins
.
Nucleic Acids Res
2019
;
47
:
3811
27
.
40.
Luo
X
,
Kraus
WL
.
On PAR with PARP: cellular stress signaling through poly(ADP-ribose) and PARP-1
.
Genes Dev
2012
;
26
:
417
32
.
41.
Chen
ZT
,
Zhao
W
,
Qu
S
,
Li
L
,
Lu
XD
,
Su
F
, et al
.
PARP-1 promotes autophagy via the AMPK/mTOR pathway in CNE-2 human nasopharyngeal carcinoma cells following ionizing radiation, while inhibition of autophagy contributes to the radiation sensitization of CNE-2 cells
.
Mol Med Rep
2015
;
12
:
1868
76
.
42.
Wyrsch
P
,
Blenn
C
,
Bader
J
,
Althaus
FR
.
Cell death and autophagy under oxidative stress: roles of poly(ADP-Ribose) polymerases and Ca2+
.
Mol Cell Biol
2012
;
32
:
3541
53
.
43.
Zhang
DX
,
Zhang
JP
,
Hu
JY
,
Huang
YS
.
The potential regulatory roles of NAD+ and its metabolism in autophagy
.
Metabolism
2016
;
65
:
454
62
.
44.
Smith
R
,
Kao
GD
.
Pictures, progress, and perplexities: the immediate cell biological effects of ionizing radiation
.
Cancer Biol Ther
2004
;
3
:
602
7
.
45.
Thorn
CF
,
Oshiro
C
,
Marsh
S
,
Hernandez-Boussard
T
,
McLeod
H
,
Klein
TE
, et al
.
Doxorubicin pathways: pharmacodynamics and adverse effects
.
Pharmacogenet Genomics
2011
;
21
:
440
6
.
46.
Castaldo
SA
,
Freitas
JR
,
Conchinha
NV
,
Madureira
PA
.
The tumorigenic roles of the cellular REDOX regulatory systems
.
Oxid Med Cell Longev
2016
;
2016
:
8413032
.
47.
Somers
K
,
Evans
K
,
Cheung
L
,
Karsa
M
,
Pritchard
T
,
Kosciolek
A
, et al
.
Effective targeting of NAMPT in patient-derived xenograft models of high-risk pediatric acute lymphoblastic leukemia
.
Leukemia
2020
;
34
:
1524
39
.
48.
Srinivas
US
,
Tan
BWQ
,
Vellayappan
BA
,
Jeyasekharan
AD
.
ROS and the DNA damage response in cancer
.
Redox Biol
2019
;
25
:
101084
.
49.
Brunyanszki
A
,
Olah
G
,
Coletta
C
,
Szczesny
B
,
Szabo
C
.
Regulation of mitochondrial poly(ADP-Ribose) polymerase activation by the β-adrenoceptor/cAMP/protein kinase a axis during oxidative stress
.
Mol Pharmacol
2014
;
86
:
450
62
.
50.
Aldini
G
,
Altomare
A
,
Baron
G
,
Vistoli
G
,
Carini
M
,
Borsani
L
, et al
.
N-Acetylcysteine as an antioxidant and disulphide breaking agent: the reasons why
.
Free Radic Res
2018
;
52
:
751
62
.
51.
Forrester
SJ
,
Kikuchi
DS
,
Hernandes
MS
,
Xu
Q
,
Griendling
KK
.
Reactive oxygen species in metabolic and inflammatory signaling
.
Circ Res
2018
;
122
:
877
902
.
52.
Naderi
S
,
Blomhoff
HK
.
Mad1 expression in the absence of differentiation: effect of cAMP on the B-lymphoid cell line Reh
.
J Cell Physiol
1999
;
178
:
76
84
.
53.
Christoffersen
J
,
Smeland
EB
,
Stokke
T
,
Taskén
K
,
Andersson
KB
,
Blomhoff
HK
.
Retinoblastoma protein is rapidly dephosphorylated by elevated cyclic adenosine monophosphate levels in human B-lymphoid cells
.
Cancer Res
1994
;
54
:
2245
50
.
54.
Gützkow
KB
,
Naderi
S
,
Blomhoff
HK
.
Forskolin-mediated G1 arrest in acute lymphoblastic leukaemia cells: phosphorylated pRB sequesters E2Fs
.
J Cell Sci
2002
;
115
:
1073
82
.
55.
Bochum
S
,
Berger
S
,
Martens
UM
.
Olaparib
.
Recent Results Cancer Res
2018
;
211
:
217
33
.
56.
Nilles
N
,
Fahrenkrog
B
.
Taking a bad turn: compromised DNA damage response in leukemia
.
Cells
2017
;
6
:
11
.
57.
Wang
H
,
Zhang
S
,
Song
L
,
Qu
M
,
Zou
Z
.
Synergistic lethality between PARP-trapping and alantolactone-induced oxidative DNA damage in homologous recombination-proficient cancer cells
.
Oncogene
2020
;
39
:
2905
20
.
58.
Trucco
C
,
Oliver
FJ
,
de Murcia
G
,
Ménissier-de Murcia
J
.
DNA repair defect in poly(ADP-ribose) polymerase-deficient cell lines
.
Nucleic Acids Res
1998
;
26
:
2644
9
.
59.
Ménissier de Murcia
J
,
Ricoul
M
,
Tartier
L
,
Niedergang
C
,
Huber
A
,
Dantzer
F
, et al
.
Functional interaction between PARP-1 and PARP-2 in chromosome stability and embryonic development in mouse
.
EMBO J
2003
;
22
:
2255
63
.

Supplementary data