Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Mol Cancer Ther. Author manuscript; available in PMC 2013 August 12.
Published in final edited form as:
PMCID: PMC3741063

Poly(ADP-ribose) polymerase inhibitor ABT-888 potentiates the cytotoxic activity of temozolomide in leukemia cells: influence of mismatch repair status and O6-methylguanine-DNA methyltransferase activity


The poly(ADP-ribose) polymerase (PARP) inhibitor ABT-888 potentiates the antitumor activity of temozolomide (TMZ). TMZ resistance results from increased O6-methylguanine-DNA methyltransferase (MGMT) activity and from mismatch repair (MMR) system mutations. We evaluated the relative importance of MGMT activity, MMR deficiency, nonhomologous end joining (NHEJ), and PARP activity in ABT-888 potentiation of TMZ. MMR-proficient and MMR-deficient leukemia cells with varying MGMT activity, as well as primary leukemia samples, were used to determine TMZ IC50 alone and with ABT-888. ABT-888 effectively inhibited PARP activity and enhanced TMZ growth inhibition in most leukemia cells. ABT-888 potentiation was most effective in MMR-deficient cells with low MGMT activity [potentiation factor (PF) = 21]. ABT-888 also potentiated TMZ activity in MMR-deficient cells with elevated MGMT activity. Unexpectedly, ABT-888 also enhanced TMZ activity in MMR-proficient cells (PF = 3–7). ABT-888 potentiation was unrelated to NHEJ activity. ABT-888 potentiated TMZ (PF = 2–5) in two of four acute myeloid leukemia patient samples but showed little potentiation in primary acute lymphoblastic leukemia. In conclusion, although ABT-888 potentiation of TMZ was most pronounced in MMR-deficient cells with low MGMT activity, neither MMR proficiency nor MGMT overexpression completely abrogated ABT-888 potentiation of TMZ.

Keywords: Acute lymphocytic leukemia, ALL, acute myeloid leukemia, AML, base excision repair, microsatellite instability, mismatch repair


DNA repair systems mediate tumor cell response to DNA-damaging anticancer agents. Temozolomide (TMZ), an alkylating agent active in the treatment of gliomas (1, 2) and some leukemias (3, 4), creates DNA damage by adding methyl adducts to N7 guanine (70% total adducts), N3 adenine (9%), and O6 guanine (5%; Fig. 1; ref. 5). The cytotoxicity of TMZ has been attributed to the creation of O6-methylguanine (6), which results in single-stranded DNA breaks, growth arrest, and apoptosis (7, 8). TMZ resistance results from at least two mechanisms: (a) increased levels of the DNA repair protein O6-methylguanine-DNA methyltransferase (MGMT), which removes O6-methyl adducts from the O6 position of guanine in DNA, and (b) deficiencies in the DNA mismatch repair (MMR) system, resulting in microsatellite instability (MSI) and tolerance of O6-methylgaunine adduct DNA mismatches (Fig. 1; ref. 9). Although MGMT inhibitors, such as O6-benzylguanine (O6-BG), can effectively overcome TMZ resistant in MMR-proficient (MSI stable) cells, they are ineffective in cells with MMR deficiencies (8, 10).

Fig. 1
Model of the effects of PARP inhibition on TMZ activity. TMZ creates O6-methylguanine adducts (CH2) and N7- and N3-methyl adducts (CH3 on either adenine or guanine), which are efficiently removed by BER. Model shows guanine as a representative nucleotide. ...

Because TMZ also generates N-methylated bases (N3 and N7), which can be removed by the base excision repair (BER) system (5), robust BER activity can result in TMZ resistance (5, 11). Central to BER and the removal of methylated N3 and N7 adducts is the enzyme poly(ADP-ribose) polymerase (PARP), an abundant nuclear enzyme that senses both single-stranded DNA and dsDNA breaks. In BER, PARP acts as a nick sensor, catalyzing the cleavage of NAD+ and attaching PARP to itself, histones, and other target proteins. Negatively charged ADP-ribose polymers create electrostatic repulsions between DNA and histones, opening chromatin for DNA repair; PARP also recruits BER proteins to sites of single-stranded DNA breaks, initiating DNA repair (12). Thus, PARP inhibitors (PARPi) may overcome TMZ resistance in MMR-deficient cells by blocking BER, resulting in cytotoxicity from N3- and N7-methyl adducts (11, 13).

PARP inhibitors have been tested in several tumor types and have been shown to enhance the antitumor effects of TMZ in leukemia (13), glioma (1416), lung (17, 18), and colon carcinoma, both in vitro (16, 1820) and in xenograft models (17, 21). Previous research has shown that the oral PARPi ABT-888 effectively inhibits PARP activity in animals (22, 23). In a phase 0 trial in humans, a single 25 mg dose of ABT-888 resulted in a median plasma ABT-888 concentration of 210 nmol/L, resulting in >92% PARP inhibition (24).

Because MMR status has been well characterized in a wide range of established leukemia cell lines, our objective was to use these cell lines as a model to assess the relative importance of MGMT activity and MMR status on the ability of ABT-888 to potentiate the growth-inhibitory effects of TMZ. ABT-888 has previously been shown to inhibit both PARP-1 and PARP-2 isoenzymes (22). Our goal was to determine (a) whether PARPi potentiation of TMZ was effective in cells with MMR proficiency, (b) whether PARPi potentiation of TMZ was abrogated by elevated MGMT, and (c) whether other mechanisms influence PARPi potentiation of TMZ.

Materials and Methods


RPMI 1640 cell culture medium, PBS, dextrose, sodium pyruvate, sodium bicarbonate, and HEPES were purchased from Life Technologies; fetal calf serum and high-glucose RPMI 1640 cell culture medium were purchased from the American Type Culture Collection; bovine growth serum was purchased from Hyclone; penicillin/streptomycin was purchased from Invitrogen; and Lymphoprep for mononuclear cell isolation was purchased from Greiner Bio-One. ABT-888 was synthesized and kindly provided by Abbott Laboratories. ABT-888 was diluted in DMSO to a stock concentration of 62 mmol/L. O6-BG (NSC 637037) was provided by the Cancer Therapy and Evaluation Program of the National Cancer Institute. TMZ (Schering-Plough) was purchased and formulated in DMSO according to the manufacturers' recommendations.

Cell Lines

The human T-cell acute lymphoblastic leukemia (ALL) cell lines Jurkat, Molt4, and HSB2; the human pre-B ALL cell lines JM1 and Reh; the B-cell lines Raji and Daudi; the histiocytic cell line U937; and the acute myeloid leukemia (AML) cell lines HL-60 (acute promyelocytic leukemia), KG1, HEL (erythroleukemia), and THP1 (monocytic leukemia) were purchased and cultured as directed by the American Type Culture Collection.

Culture of Primary Leukemia Cells

Leukemia cells were obtained from peripheral blood, leukapheresis, or bone marrow aspirate specimens from children with newly diagnosed acute leukemia before chemotherapy in accordance with Institutional Review Board guidelines. Peripheral blood mononuclear cells were isolated using Lymphoprep and frozen at a cell density of 1 × 107/mL at −80°C until use. Primary leukemia cells were cultured in RPMI 1640 supplemented with 20% FCS and penicillin/streptomycin. During drug sensitivity assays, cell viability was determined by trypan blue exclusion at 48 or 72 h and noted to be >90% in the absence of drug.

In vitro Cytotoxicity Assays

The growth inhibition effect of ABT-888 and TMZ was assessed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) colorimetric dye reduction as previously described (25) or the CellTiter-Glo luminescent cell viability assay (Promega) according to the manufacturer's instructions. Leukemia cell lines were plated at a cell density of 0.5 to 2 × 105/mL. TMZ was serially diluted over a 106-fold range of concentrations to determine single-agent IC50s as described (4). For assays assessing single drug activity or TMZ in combination with ABT-888, replicates of six-wells were used for each drug concentration and the assay was repeated using two replicate plates. ABT-888 was tested in MMR-proficient U937, THP1, and JM1 and in MMR-deficient HSB2, Molt4, Jurkat, and Reh (Table 1). IC50 values for each cell line were determined in at least three independent experiments using the Hill equation as previously described (25). Primary cells were plated at a concentration of 8 to 40 × 104/mL and drug concentrations were tested in triplicate. IC50 values for each primary cell sample were determined in at least two independent experiments. In combination experiments, ABT-888 and/or O6-BG were added 30 min before TMZ. Viability was assessed after 72 h. Representative cell lines combining TMZ with ABT-888 are shown from each MGMT and MSI subgroup (Fig. 3A and B). Combinations between TMZ, O6-BG, and ABT-888 were tested in the MMR-proficient cell lines JM1 and U937 as well as the MMR-deficient cell lines Jurkat and Molt4. The combination effect of drugs used with TMZ (O6-BG and ABT-888) was modeled using the universal response surface approach method as described (25, 26).

Fig. 3
ABT-888 potentiation of TMZ in MMR-proficient (A) or MMR-deficient (B) cell lines. A, AML cell lines THP1 (S1, left), HL-60 (S2, middle), and KG1 (S3, right) treated with TMZ alone (○) or TMZ in combination with 0.5 µmol/L (●) ...
Table 1
Summary of leukemia subtype, MGMT activity, and MSI status in leukemia cell lines

MGMT Activity, MSI, and PARP Activity Assays

MGMT activity in peripheral blood mononuclear cell or bone marrow aspirate tumor cell lysates was determined by the removal of O6-[3H]methylguanine from a 3H-methylated DNA substrate and quantified as fmol O6-[3H]methylguanine/mg total protein as described (27). For MSI determination in four representative primary leukemia cells (patients p115 and p120 with AML and patients p152 and p157 with ALL), genomic DNA was extracted and amplified using three MSI multiplex reaction mixtures containing National Cancer Institute panel markers (BAT-25, BAT-26, D2S123, D17S250, and D5S346; ref. 28), quasimonomorphic mononucleotide markers (NR-21, NR-22, NR-24, BAT-25, and BAT-26; ref. 29), or an alternative panel of markers (D18S35, TP53-DI, TP53-PENTA, D1S2883, and FGA) as described (4). MSI status of established cell lines was obtained from the literature (9, 3033). PARP activity was measured by the incorporation of biotinylated poly(ADP-ribose) into histones using modifications of a commercially available PARP assay (Trevigen). Cell lysates were prepared from 1 × 107 leukemia cells using 1× PARP buffer (Trevigen) supplemented with 0.4 mmol/L phenylmethylsulfonyl fluoride, one-half complete protease inhibitor cocktail tablet (Roche), 1% NP40, and 0.1% SDS. Each sample was done in triplicate, and recombinant PARP (Trevigen) was used to produce a standard curve for each assay. PARP activity was assessed in either duplicate or triplicate in at least two independent experiments. PARP inhibition with ABT-888 was determined in MMR-proficient cell lines JM1 and U937 as well as MMR-deficient cell lines Molt4, Reh, Jurkat, and HSB2.


Cell lysates were prepared from 1 × 107 cells as previously described (34). Thirty to 50 µg of protein extract from representative cell lines (MMR-proficient cell lines JM1, U937, and THP1 as well as MMR-deficient cell lines Molt4, Reh, HSB2, and Jurkat) were run on SDS-PAGE gels and transferred to nitrocellulose membranes (Bio-Rad). Membranes were incubated overnight with MGMT (1:500 dilution; Cell Signaling Technology), 1 µg/mL anti-human PARP (BD Pharmingen), 4 µg/mL anti-Ku70 (clone N3H10; Abcam), rabbit anti-human phospho-histone H2AX (1:1,000 dilution; Cell Signaling Technology), or β-actin (1:10,000 dilution; Sigma) antibodies diluted in Odyssey blocking buffer (Li-Cor). Bound primary antibodies were detected with IR-800 or IR-700 dye-labeled, appropriate species-specific secondary antisera and visualized on a Li-Cor Odyssey IR scanner. The intensity of gel bands was measured using Molecular Dynamics ImageQuant software (version 5.2).

Nonhomologous End-Joining Assay

Small-scale in vitro nonhomologous end-joining (NHEJ) assays were done essentially as described (35), except that a 2.9-kb PvuII- and HindIII-linearized 5′-32P-labeled pCDNA3.1 fragment was used. Reactions (10 µL) were done with 30 µg of protein extract and 10 ng of 32P-labeled DNA. Full image is provided in Supplementary Fig. S1.

Transient Transfection Assay

pCDNA3.3 and pcDNA-MGMT plasmids were transfected into MMR-proficient THP1 AML cells and MMR-deficient HSB2 T-cell ALL cells using Nucleofector Kit V from Amaxa, Inc. as directed with minor modifications. Approximately 1.5 × 106 cells were suspended in 100 µL of Nucleofector Solution V containing 0.5 µg of plasmid DNA. Plasmid pEGFP was used to monitor transfection efficiency. Following 6 to 8 h of incubation at 37°C, cells were washed and cultured for another 40 h before protein lysates were made and cytotoxicity assays were done. The transfection efficiency, measured by counting the green fluorescent cells, was estimated to be 70% to 80% for each of the cell lines used (full image in Supplementary Fig. S1)


Dose-response curves are shown using mean ± SD. IC50s were calculated using the Hill equation as previously described (25). MGMT activity and PARP activity are reported as mean ± SD of each sample done in triplicate. PARP activity in MMR-proficient and MMR-deficient cell lines was compared using a two-sided Student's t test. PARP activity, MGMT activity, and potentiation factors (PF) were compared as continuous variables using the Wilcoxon rank-sum test. K50s and PF were determined from at least three independent experiments.


TMZ and PARP Inhibition in Leukemia Cell Lines

The growth-inhibitory effects of TMZ and the PARPi ABT-888 were tested in MMR-proficient and MMR-deficient leukemia cell lines with variable MGMT activity (Table 1). Previous work has shown that plasma TMZ concentrations (Cmax) reached ~62 µmol/L when given at the maximum tolerated dose (200 mg/m2) over 5 days (36). As shown in Fig. 2A, most leukemia cell lines were resistant to TMZ at this concentration. Although three MMR-proficient cell lines (JM1, HEL, and U937) were more sensitive to TMZ (IC50, 11–50 µmol/L; Fig. 2A), MMR-deficient cell lines were resistant to TMZ, with an average IC50 of 450 µmol/L. Leukemia cells with elevated MGMT activity (Jurkat, Reh, and KG1) were also more resistant to TMZ than cell lines with absent (U937 and HEL) or low (JM1) MGMT activity.

Fig. 2
TMZ and ABT-888 growth inhibition in leukemia cell lines. MTT cytotoxicity assays were done with either TMZ (A) or ABT-888 (B) at concentrations ranging from 1 to 2,000 µmol/L. Percentage cell survival is plotted as a function of drug concentration. ...

ABT-888 also inhibited leukemia cell growth, with IC50s ranging from 20 to 196 µmol/L (Fig. 2B). These ABT-888 inhibitory concentrations, however, were approximately 5- to 33-fold higher than plasma concentrations achieved in either animals or humans (22, 24). Because steady-state concentrations in animals ranged from 0.35 to 1 µmol/L, and the single-dose ABT-888 Cmax was 0.2 to 11 µmol/L, further exploration of the ABT-888–potentiating effects was done using concentrations of 0.5 and 5 µmol/L.

Variable MGMT Activity and MMR Status in Leukemia Cell Lines

MGMT expression is variable in primary leukemia cells, ranging from undetectable to 6,000 fmol/mg protein (4). As shown in Table 1, MGMT activity varied in both MMR-proficient (MSI stable) and MMR-deficient (MSI unstable) leukemia cell lines. In MMR-proficient cells, MGMT activity varied from undetectable in the AML cell lines HEL and U937 to high (1,170 fmol/mg protein) in the AML cell line KG1. In MMR-deficient (MSI unstable) cell lines, MGMT activity ranged from 320 fmol/mg protein in the T-cell ALL line HSB2 to 1,300 fmol/mg protein in the pre-B ALL cell line Reh. To assess the relative contributions of MGMT activity and MMR status to ABT-888 potentiation of TMZ, leukemia cell lines were subdivided into seven groups based on relative MGMT activity and MMR status (Table 1). Because normal adult peripheral blood mononuclear cells express an average of 770 + 170 fmol/mg protein MGMT activity (4), MMR-proficient cell lines (MSI stable; S) were divided into those with undetectable (S0), low (S1), average (S2), or high (S3) MGMT activity, corresponding to undetectable, <460 fmol/mg, 460 to 1,100 fmol/mg, and >1,110 fmol/mg MGMT protein, respectively. MMR-deficient cell lines (MSI unstable; U) were similarly classified into groups with low (U1), average (U2), and high (U3) MGMT activity based on the same criteria.

ABT-888 Potentiates TMZ Activity in Cells with Proficient MMR and MGMT Activity

With one exception (U937 AML cells), ABT-888 enhanced the activity of TMZ in MMR-proficient cells from 3- to 7-fold, an effect that seemed to be independent of MGMT activity (Fig. 3A and C). ABT-888 was more effective in potentiating TMZ-induced cytotoxicity in MMR-deficient cells (10- to 21-fold; Fig. 3B) and was most potent in MMR-deficient leukemia cells with low MGMT activity (U1), such as the T-cell ALL line HSB2 (Fig. 3B, left). In this cell line, ABT-888 decreased the TMZ IC50 from 440 to 20 µmol/L, a 21-fold enhancement. ABT-888 alone had no effect on HSB2 growth at the same concentration (5 µmol/L; Fig. 2B). ABT-888 also enhanced the effects of TMZ in MMR-deficient cell lines with average or elevated MGMT activity. In pre-B and T-ALL cell lines with the highest MGMT activity (U3), ABT-888 provided a 10- to 13-fold potentiation (Fig. 3B, right). This suggests that ABT-888 can enhance TMZ-induced cytotoxicity in MMR-deficient leukemia cell lines despite elevated MGMT activity.

Combination of MGMT Inhibitor O6-BG with ABT-888 and TMZ

Because MGMT inhibitors have been extensively studied as a means to increase TMZ efficacy (37, 38), it was of interest to determine the relative potentiation of the MGMT inhibitor O6-BG and ABT-888. In MMR-proficient cell lines, such as JM1, O6-BG seemed to provide more TMZ potentiation than ABT-888 (Fig. 3D) and ABT-888 did not provide any additional TMZ potentiation. In MMR-deficient cell lines, however, such as the T-cell ALL cell lines Molt4 (Fig. 3E) and Jurkat (data not shown), the addition of both O6-BG and ABT-888 each potentiated TMZ and the addition of both O6-BG and ABT-888 to TMZ was additive when modeled using the universal response surface approach method (26).

Overexpression of MGMT Does Not Prevent ABT-888 Potentiation of TMZ

To determine if MGMT overexpression could overcome ABT-888 potentiation of TMZ growth inhibition, we transfected the MMR-proficient THP1 and MMR-deficient HSB2 cells with a construct constitutively expressing MGMT. By immunoblot densitometry, MGMT expression was increased at least 30-fold (Fig. 4A). Increase in MGMT expression was able to partially, but not completely, overcome ABT-888 potentiation of TMZ cytotoxicity. In the MMR-proficient THP1 cells (Fig. 4B), PF decreased from 3.4 to 1.6, and in the MMR-deficient cell lines HSB2 (Fig. 4C), the PF decreased from 15 to 5.8.

Fig. 4
Effect of MGMT overexpression on ABT-888 potentiation of TMZ in MMR-proficient and MMR-deficient leukemia cell lines. A, confirmation of increased MGMT expression by Western blot and quantitation of increased MGMT expression by densitometry. The MMR-proficient ...

ABT-888 Potentiation of TMZ: Relationship to PARP Activity

To determine whether ABT-888–mediated TMZ potentiation was related to the degree of PARP inhibition, immunoblots and activity assays were done in several of the leukemia cell lines (Fig. 5). PARP protein (Fig. 5A) and PARP activity levels (Fig. 5B) in MMR-proficient and MMR-deficient cell lines were quite variable and did not seem to correlate with MGMT activity. MMR-proficient cell lines had PARP activity ranging from 45 fmol/µg protein (AML cells) to 3,430 fmol/µg protein (HEL AML cells). Although leukemia cells with MMR deficiencies (MSI unstable) tended to have higher PARP activity (510–2,050 fmol/µg protein) than MMR-proficient cells, the difference between MMR-proficient and MMR-deficient leukemia cell lines was not statistically significant (P = 0.76, Student's t test).

Fig. 5
Mechanisms of PARP potentiation in leukemia cell lines. A, pretreatment PARP, Ku70, and phospho-H2AX protein expression in four MMR-deficient (U) and three MMR-proficient (S) leukemia cell lines. Expression is normalized to actin expression. B, PARP activity ...

Overall, ABT-888 inhibited PARP activity by 78% to 97% (Fig. 5B). The degree of PARP inhibition was roughly proportional to the degree of ABT-888 potentiation of TMZ in most leukemia cell lines. For example, ABT-888 induced 90% PARP inhibition in the MMR-deficient cell line HSB2, which showed the greatest degree of ABT-888–induced TMZ enhancement (21-fold). Residual PARP activity in HSB2 was only 60 fmol/µg protein. In contrast, in the MMR-proficient JM1, which showed less potentiation (3-fold), ABT-888 provided only 78% inhibition of PARP activity and the highest residual PARP activity (360 fmol/µg protein) was observed.

Notably, ABT-888 did not potentiate TMZ growth inhibition in the AML cell line U937, a MMR-proficient cell line with minimal MGMT activity (Fig. 5D). The U937 cell line expressed very little PARP by immunoblot (Fig. 5A) and had minimal PARP activity (Fig. 5B), suggesting that ABT-888 may be less effective at potentiating TMZ in cells with very low PARP activity.

In summary, when MGMT activity, PARP activity, and the PFs were examined as continuous variables, there were no significant differences in MGMT activity or PARP activity between the MMR-deficient and the MMR-proficient cell lines (P = 0.37 and 0.23, respectively, Wilcoxon rank-sum test). PFs, however, were significantly higher in MMR-deficient cell lines when compared with those with intact MMR (P = 0.02, Wilcoxon rank-sum test).

ABT-888 Potentiation of TMZ Is Not Due to NHEJ

PARP also participates in forms of DNA double-strand break (DSB) repair, including both homologous and nonhomologous recombination and a backup pathway of NHEJ in cells compromised for classic NHEJ (12, 39, 40). To determine whether PARP potentiation of TMZ was affected by NHEJ in leukemia cell lines, we examined the expression of Ku70, which was strongly expressed in all leukemia cell lines tested, including U937 (Fig. 5A).

To further determine whether PARPi potentiation of TMZ was affected by NHEJ status, we measured NHEJ activity in leukemia cell lines using an in vitro NHEJ assay (35). Leukemia cell lysates were incubated with a radiolabeled probe designed to dimerize as the results of NHEJ as shown in 293T cells (Fig. 5E). MMR-proficient cell lines with similar ABT-888 potentiation of TMZ had different NHEJ activity levels. The AML cell lines THP1 and HL-60 had undetectable NHEJ, whereas in the pre-B ALL JM1 cell line NHEJ was robust. We also examined phosphorylated histone H2AX, which is increased in response to unrepaired DSBs, and found no correlation between H2AX protein and ABT-888 potentiation of TMZ (Fig. 5A). These data suggest that ABT-888 potentiation of TMZ is not related to its role in NHEJ.

ABT-888 Potentiation of TMZ in Primary Leukemia Cells Is Related to Leukemia Subtype

Because leukemia cell lines have developed multiple adaptations to growth in culture, we also examined the ability of ABT-888 to potentiate TMZ in primary acute leukemia cells obtained directly from patients (Fig. 6). We determined whether ABT-888 potentiation correlated with MGMT activity or, in a subset of patients, with MMR status (Table 2). ABT-888 did not significantly potentiate TMZ activity in primary ALL leukemia cells (Fig. 6A; Table 2) but enhanced the growth-inhibitory effects of TMZ in two of four AML leukemias (Fig. 6B; Table 2). The lack of ABT-888 potentiation of TMZ in two primary AML specimens did not seem to correlate with MGMT expression, as patient specimen p115, which showed the greatest ABT-888–induced potentiation of TMZ among the primary leukemias (4.6-fold), had elevated MGMT activity (2,580 fmol/mg protein). All tested primary leukemia cells were MMR proficient.

Fig. 6
PARP activity and ABT-888 potentiation of TMZ in primary leukemia samples. A to D, dose-response curve of TMZ alone (○) or TMZ in combination with either 0.5 µmol/L (●) or 5 µmol/L ([filled square]) of ABT-888 in four representative ...
Table 2
Summary of MGMT activity, PARP activity, and ABT-888 potentiation of TMZ in primary pediatric cells

To determine whether ABT-888 potentiation of TMZ was related to baseline PARP activity, we measured PARP activity in 14 primary leukemia samples (Fig. 6C). Similar to leukemia cell lines, primary leukemia cells had widely variable PARP activity, ranging from <40 fmol/µg protein (below the level of quantitation) to 1,860 fmol/µg protein. There seemed to be little correlation between PARP activity and ABT-888 potentiation of TMZ. Of interest, one of the two AML primary samples that showed no ABT-888 potentiation of TMZ (sample 4) also showed no PARP activity similar to the U937 cell line.


In this study, we used a panel of leukemia cell lines to assess the relative effects of MGMT activity, MMR status, and PARP activity on the ability of ABT-888 to potentiate the antitumor effects of TMZ. Although our data indicate that ABT-888 was most effective in potentiating TMZ in MMR-deficient cell lines with low MGMT activity, ABT-888 also potentiated TMZ in MMR-proficient cell lines. In primary acute leukemia cells, ABT-888 potentiation of TMZ seemed to correlate only with leukemia subtype, providing >3-fold potentiation in two of four primary AML samples.

Although Tentori et al. (41) had also noted that leukemia cell lines could be sensitized to TMZ using the PARPi benzamine, this is the first report to examine the effect of MGMT activity on the ability of PARPis to potentiate TMZ in leukemia MMR-proficient and MMR-deficient cell lines. This is also the first reported examination of ABT-888 potentiation of TMZ in primary acute leukemia. Similar to previous reports in leukemia cell lines, colon cancer, and glioma, PARPi potentiation of TMZ was more pronounced in cells with MMR deficiencies (11, 14, 19).

Several PARPis are under investigation in oncology clinical trials. ABT-888 is undergoing phase 1 testing in adults with melanoma. Other PARPis under clinical study include AZD2281 (AstraZeneca), AG014699 (Pfizer) in combination with TMZ for adults with solid tumors and metastatic malignant melanoma (42), INO-1001 (Inotek/Genentech) in combination with TMZ for metastatic melanoma and glioblastoma multiforme (43), KU59436 (KuDOS/AstraZeneca) in advanced solid tumors (including patients with BRCA1/BRCA2 mutations; ref. 44), and BS201 (BiPar) for refractory solid tumors and lymphomas (12). A better understanding of the factors that govern PARPi potentiation of TMZ may allow for the prospective determination of which tumors most likely to respond to this combination.

Prior clinical studies in both adults and pediatrics have suggested that TMZ may be effective in AML patients with low/absent MGMT tumor activity but is less effective in the presence of MMR deficiency or elevated MGMT activity (3, 4). Our data suggest that PARPi may be effective at potentiating TMZ in AML patients without MMR deficiencies and that this potentiation may be independent of MGMT activity. Although common in adults with treatment-related AML (45), MMR deficiencies are uncommon in pediatric patients with newly diagnosed AML. However, pediatric patients with relapsed leukemia frequently have elevated MGMT (4) and this group of patients would benefit from potentiating agents that could overcome both MMR deficiencies and elevated MGMT activity.

In MMR-deficient cell lines, we found an inverse correlation between MGMT activity and potentiation of TMZ by ABT-888. MGMT activity seemed less important in determining ABT-888 potentiation of TMZ than MMR status in these leukemia cell lines. First, we observed 3-fold ABT-888 potentiation of TMZ in MMR-proficient leukemia cell lines and 10-fold ABT-888 potentiation of TMZ in MMR-deficient leukemia cells with elevated MGMT activity (Fig. 3C). Second, overexpression of MGMT in both MMR-deficient and MMR-proficient cell lines was unable to completely block ABT-888 potentiation of TMZ (Fig. 4). PARPi potentiation of TMZ in the presence of MGMT was also observed using the PARPi 3AB in colon cancer cell lines (46). These data suggest that neither increased MGMT activity nor MMR proficiency precludes the ability of PARPi to potentiate TMZ.

In some cases, the effects of ABT-888 seemed to depend on a threshold level of PARP activity because ABT-888 did not potentiate TMZ in either the U937 AML cell line or primary leukemia cells with very low PARP activity (Figs. 5C and and6A).6A). However, ABT-888 did not result in TMZ potentiation in many primary ALL samples with high PARP activity, indicating that the presence of robust PARP activity is not sufficient for ABT-888 potentiation. There was also one AML cell line (KG1) with very low PARP activity, which showed 3-fold ABT-888 potentiation of TMZ, suggesting that either ABT-888 potentiation of TMZ may be dependent on nonribosylation functions of PARP. TMZ potentiation effects that are independent of PARP inhibition (i.e., off-target effects) cannot be excluded. Previous data showed that the PARPi GPI-15427 increased the antitumor activity of TMZ and irinotecan in colon cancer xenografts irrespective of PARP activity (16). Although the range of PARP activity in most primary tumors is unknown, Plummer et al. (47) examined PARP activity in adult melanoma specimens and noted that PARP activity was quite variable, ranging from 1.7 to 3,600 pmol PARP/mg protein. Hence, it is likely that the effectiveness of PARPi will vary, reflecting variability in both baseline PARP activity and the efficiency of PARP inhibition.

Although best known for its role in BER, PARP also participates in DSB repair (5). Evidence suggests that PARP competes with Ku for DSBs and PARP can either preserve DSBs for homologous recombination (48) or (in the absence of Ku) initiate the use of a backup NHEJ repair pathway involving DNA ligase III (40). The effects of PARP inhibition could be influenced by the intactness of the DSB repair pathways; for instance, in the absence of Ku or BRCA1/2, tumor cells may be very sensitive to PARP inhibition (49). We have shown here that NHEJ status did not affect the ability of ABT-888 to potentiate TMZ in leukemia cell lines. Further study is needed to determine whether the status of the backup NHEJ repair pathway or homologous recombination influences ABT-888 potentiation of TMZ.

The influence of MGMT activity, PARP activity, and MSI on the ability of ABT-888 to potentiate TMZ is derived from a limited number of leukemia cell lines and therefore should be interpreted with caution. PARPi potentiation of TMZ in MMR-proficient cell lines may also be limited clinically because ABT-888 potentiation is seen only at high TMZ concentrations (>100 µmol/L). PARPi potentiation of TMZ could also be influenced by factors that were not examined, such as ABT-888 transport or metabolism.

In limited phase 1 studies of TMZ in patients with leukemia, where patients may have either ALL or AML, clinical responses have been observed exclusively in patients with AML. In a phase 1 study of TMZ in pediatric leukemia (4), two of eight patients with AML had objective responses. These patients had absent or low MGMT activity and no MSI. Similarly, in an adult phase 1 study of TMZ in acute leukemia, the four patients with an objective response to TMZ (complete remission or complete remission with incomplete platelet recovery) had AML (3). In this study, in vitro ABT-888 potentiation of TMZ was seen in some primary samples from patients with AML (Table 2). Although the reason for this is unclear, it may be that more complete PARP inhibition can be obtained in patients with AML.

In summary, we have shown that the PARPi ABT-888 can effectively enhance the activity of TMZ, potentiating TMZ growth inhibition 3- to 21-fold in leukemia cell lines. ABT-888 potentiation of TMZ was most effective in cells with MMR deficiencies and low MGMT activity. Unexpectedly, PARP inhibition also potentiated TMZ activity in MMR-proficient leukemia cell lines, providing a 3- to 7-fold enhancement of TMZ growth inhibition independent of MGMT activity. Our data suggest that ABT-888 may function as a useful potentiator of TMZ; further studies, particularly in AML, are warranted.

Supplementary Material



We thank Gouqing (Mary) Ge for excellent technical assistance and Meng-Fen Wu for statistical support.

Grant support: K12 CA90433-04 (T.M. Horton), K23 CA113775 (T.M. Horton), Scott Carter NCCF Research Fellowship (T.M. Horton), Ladies Leukemia League (T.M. Horton), RO1 CA109478 (D. Pati), RO1 CA81485 (M.E. Dolan), and U01 CA63187 (University of Chicago Cancer Research Center). Pharmacologic studies were supported by the University of Chicago Cancer Research Center Pharmacology Core Facility ( through the University of Chicago Cancer Research Center Cancer Center Support Grant P30 CA14599.


Disclosure of Potential Conflicts of Interest

M.E. Dolan: coinventor of O6-BG, which has been licensed to Access Oncology, Inc. No other potential conflicts of interest were disclosed.

Note: Supplementary material for this article is available at Molecular Cancer Therapeutics Online (


1. Friedman HS, Kerby T, Calvert H. Temozolomide and treatment of malignant glioma. Clin Cancer Res. 2000;6:2585–2597. [PubMed]
1. Kuo DJ, Weiner HL, Wisoff J, Miller DC, Knopp EA, Finlay JL. Temozolomide is active in childhood, progressive, unresectable, low-grade gliomas. J Pediatr Hematol Oncol. 2003;25:372–378. [PubMed]
2. Seiter K, Liu D, Loughran T, Siddiqui A, Baskind P, Ahmed T. Phase I study of temozolomide in relapsed/refractory acute leukemia. J Clin Oncol. 2002;20:3249–3253. [PubMed]
3. Horton TM, Thompson PA, Berg SL, et al. Phase I pharmacokinetic and pharmacodynamic study of temozolomide in pediatric patients with refractory or recurrent leukemia: a Children's Oncology Group Study. J Clin Oncol. 2007;25:4922–4928. [PubMed]
4. Trivedi RN, Almeida KH, Fornsaglio JL, Schamus S, Sobol RW. The role of base excision repair in the sensitivity and resistance to temozolomide-mediated cell death. Cancer Res. 2005;65:6394–6400. [PubMed]
5. Gerson SL, Miller K, Berger NA. O6 alkylguanine-DNA alkyltransferase activity in human myeloid cells. J Clin Invest. 1985;76:2106–2114. [PMC free article] [PubMed]
6. Lewis C, Low JA. Clinical poly(ADP-ribose) polymerase inhibitors for the treatment of cancer. Curr Opin Investig Drugs. 2007;8:1051–1056. [PubMed]
7. Tentori L, Graziani G, Gilberti S, Lacal PM, Bonmassar E, D'Atri S. Triazene compounds induce apoptosis in O6-alkylguanine-DNA alkyltransferase deficient leukemia cell lines. Leukemia. 1995;9:1888–1895. [PubMed]
8. Gu L, Cline-Brown B, Zhang F, Qiu L, Li GM. Mismatch repair deficiency in hematological malignancies with microsatellite instability. Oncogene. 2002;21:5758–5764. [PubMed]
9. Liu L, Markowitz S, Gerson SL. Mismatch repair mutations override alkyltransferase in conferring resistance to temozolomide but not to 1,3-bis(2-chloroethyl)nitrosourea. Cancer Res. 1996;56:5375–5379. [PubMed]
10. Liu L, Taverna P, Whitacre CM, Chatterjee S, Gerson SL. Pharmacologic disruption of base excision repair sensitizes mismatch repair-deficient and -proficient colon cancer cells to methylating agents. Clin Cancer Res. 1999;5:2908–2917. [PubMed]
11. Ratnam K, Low JA. Current development of clinical inhibitors of poly(ADP-ribose) polymerase in oncology. Clin Cancer Res. 2007;13:1383–1388. [PubMed]
12. Tentori L, Portarena I, Vernole P. Effects of single or split exposure of leukemic cells to temozolomide, combined with poly(ADP-ribose) polymerase inhibitors on cell growth, chromosomal aberrations and base excision repair components. Cancer Chemother Pharmacol. 2001;47:361–369. [PubMed]
13. Tentori L, Portarena I, Torino F, Scerrati M, Navarra P, Graziani G. Poly(ADP-ribose) polymerase inhibitor increases growth inhibition and reduces G(2)/M cell accumulation induced by temozolomide in malignant glioma cells. Glia. 2002;40:44–54. [PubMed]
14. Cheng CL, Johnson SP, Keir ST, et al. Poly(ADP-ribose) polymerase-1 inhibition reverses temozolomide resistance in a DNA mismatch repair-deficient malignant glioma xenograft. Mol Cancer Ther. 2005;4:1364–1368. [PubMed]
15. Tentori L, Leonetti C, Scarsella M, et al. Systemic administration of GPI 15427, a novel poly(ADP-ribose) polymerase-1 inhibitor, increases the antitumor activity of temozolomide against intracranial melanoma, glioma and lymphomas. Clin Cancer Res. 2003;9:5370–5379. [PubMed]
16. Miknyoczki SJ, Jones-Bolin S, Pritchard S. Chemopotentiation of temozolomide, irinotecan, and cisplatin activity by CEP-6800, a poly(ADP-ribose) polymerase inhibitor. Mol Cancer Ther. 2003;2:371–382. [PubMed]
17. Calabrese CR, Almassy R, Barton S, et al. Anticancer chemosensitization and radiosensitization by the novel poly(ADP-ribose) polymerase-1 inhibitor AG14361. J Natl Cancer Inst. 2004;96:56–67. [PubMed]
18. Curtin NJ, Wang LZ, Yiakouvaki A, et al. Novel poly(ADP-ribose) polymerase-1 inhibitor, AG14361, restores sensitivity to temozolomide in mismatch repair-deficient cells. Clin Cancer Res. 2004;10:881–889. [PubMed]
19. Calabrese CR, Batey MA, Thomas HD, et al. Identification of potent nontoxic poly(ADP-ribose) polymerase-1 inhibitors: chemopotentiation and pharmacological studies. Clin Cancer Res. 2003;9:2711–2718. [PubMed]
20. Tentori L, Leonetti C, Scarsella M, et al. Combined treatment with temozolomide and poly(ADP-ribose) polymerase inhibitor enhances survival of mice bearing hematologic malignancy at the central nervous system site. Blood. 2002;99:2241–2244. [PubMed]
21. Donawho CK, Luo Y, Penning TD, et al. ABT-888, an orally active poly(ADP-ribose) polymerase inhibitor that potentiates DNA-damaging agents in preclinical tumor models. Clin Cancer Res. 2007;13:2728–2737. [PubMed]
22. Albert JM, Cao C, Kim KW, et al. Inhibition of poly(ADP-ribose) polymerase enhances cell death and improves tumor growth delay in irradiated lung cancer models. Clin Cancer Res. 2007;13:3033–3042. [PubMed]
23. Kummar S, Kinders R, Gutierrez M, et al. Inhibition of poly (ADP-ribose) polymerase (PARP) by ABT-888 in patients with advanced malignancies: results of a phase 0 trial. J Clin Oncol. 2007;25:3518a. [PubMed]
24. Horton TM, Gannavarapu A, Blaney SM, D'Argenio DZ, Plon SE, Berg SL. Bortezomib interactions with chemotherapy agents in acute leukemia in vitro. Cancer Chemother Pharmacol. 2006;58:13–23. [PubMed]
25. Greco WR, Park HS, Rustum YM. Application of a new approach for the quantitation of drug synergism to the combination of cis-diamminedichloroplatinum and 1-β-D-arabinofuranosylcytosine. Cancer Res. 1990;50:5318–5327. [PubMed]
26. Dolan ME, Roy SK, Fasanmade AA, Paras PR, Schilsky RL, Ratain MJ. O6-benzylguanine in humans: metabolic, pharmacokinetic, and pharmacodynamic findings. J Clin Oncol. 1998;16:1803–1810. [PubMed]
27. Boland CR, Thibodeau SN, Hamilton SR, et al. A National Cancer Institute Workshop on Microsatellite Instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res. 1998;58:5248–5257. [PubMed]
28. Suraweera N. Evaluation of tumor microsatellite instability using five quasimonomorphic mononucleotide repeats. Gastroenterology. 2002;123:1804–1811. [PubMed]
29. Taverna P, Liu L, Hanson AJ, Monks A, Gerson SL. Characterization of MLH1 and MSH2 DNA mismatch repair proteins in cell lines of the NCI anticancer drug screen. Cancer Chemother Pharmacol. 2000;46:507–516. [PubMed]
30. Hangaishi A, Ogawa S, Mitani K, et al. Mutations and loss of expression of a mismatch repair gene, hMLH1, in leukemia and lymphoma cell lines. Blood. 1997;89:1740–1747. [PubMed]
31. Ribeiro EM, Rodriguez JM, Coser VM. Microsatellite instability and cytogenetic survey in myeloid leukemias. Braz J Med Biol Res. 2002;35:153–159. [PubMed]
32. Hosoya N, Hangaishi A, Ogawa S, et al. Frameshift mutations of the hMSH6 gene in human leukemia cell lines. Jpn J Cancer Res. 1998;89:33–39. [PubMed]
33. Pati D, Haddad BR, Haegele A, et al. Hormone-induced chromosomal instability in p53-null mammary epithelium. Cancer Res. 2004;64:5608–5616. [PubMed]
34. Smeaton MB, Miller PS, Ketner G, Hanakahi LA. Small-scale extracts for the study of nucleotide excision repair and non-homologous end joining. Nucleic Acids Res. 2007;35:e152. Epub 2007 Dec 10. [PMC free article] [PubMed]
35. Dhodapkar M, Rubin J, Reid JM. Phase I trial of temozolomide (NSC 362856) in patients with advanced cancer. Clin Cancer Res. 1997;3:1093–1100. [PubMed]
36. Quinn JA, Desjardins A, Weingart J, et al. Phase I trial of temozolomide plus O6-benzylguanine for patients with recurrent or progressive malignant glioma. J Clin Oncol. 2005;23:7178–7187. [PubMed]
37. Warren KE, Aikin AA, Libucha M, et al. Phase I study of O6-benzylguanine and temozolomide administered daily for 5 days to pediatric patients with solid tumors. J Clin Oncol. 2005;23:7646–7653. [PubMed]
38. Jagtap P, Szabo C. Poly(ADP-ribose) polymerase and the therapeutic effects of its inhibitors. Nat Rev Drug Discov. 2005;4:421–440. [PubMed]
39. Wang M, Wu W, Rosidi B, et al. PARP-1 and Ku compete for repair of DNA double strand breaks by distinct NHEJ pathways. Nucleic Acids Res. 2006;34:6170–6182. [PMC free article] [PubMed]
40. Tentori L, Lacal PM, Benincasa E, et al. Role of wild-type p53 on the antineoplastic activity of temozolomide alone or combined with inhibitors of poly(ADP-ribose) polymerase. J Pharmacol Exp Ther. 1998;285:884–893. [PubMed]
41. Thomas HD, Calabrese CR, Batey MA, et al. Preclinical selection of a novel poly(ADP-ribose) polymerase inhibitor for clinical trial. Mol Cancer Ther. 2007;6:945–956. [PubMed]
42. Mason KA, Valdecanas D, Hunter NR, Milas L. INO-1001, a novel inhibitor of poly(ADP-ribose) polymerase, enhances tumor response to doxorubicin. Invest New Drugs. 2008;26:1–5. [PubMed]
43. Curtin NJ. PARP inhibitors for cancer therapy. Expert Rev Mol Med. 2005;7:1–20. [PubMed]
44. Das-Gupta EP, Seedhouse CH, Russell NH. Microsatellite instability occurs in defined subsets of patients with acute myeloblastic leukaemia. Br J Haematol. 2001;114:307–312. [PubMed]
45. Wedge SR, Porteous JK, Newlands ES. 3-Aminobenzamide and/or O6-benzylguanine evaluated as an adjuvant to temozolomide or BCNU treatment in cell lines of variable mismatch repair status and O6-alkylguanine-DNA alkyltransferase activity. Br J Cancer. 1996;74:1030–1036. [PMC free article] [PubMed]
46. Plummer ER, Middleton MR, Jones C, et al. Temozolomide pharmacodynamics in patients with metastatic melanoma: DNA damage and activity of repair enzymes O6-alkylguanine alkyltransferase and poly(ADP-ribose) polymerase-1. Clin Cancer Res. 2005;11:3402–3409. [PubMed]
47. Hochegger H, Dejsuphong D, Fukushima T, et al. Parp-1 protects homologous recombination from interference by Ku and ligase IV in vertebrate cells. EMBO J. 2006;25:1305–1314. [PubMed]
48. McCabe N, Turner NC, Lord CJ, et al. Deficiency in the repair of DNA damage by homologous recombination and sensitivity to poly(ADP-ribose) polymerase inhibition. Cancer Res. 2006;66:8109–8115. [PubMed]