Purpose: To assess efficacy of the novel, selective poly(ADP-ribose) polymerase-1 (PARP-1) inhibitor AZD2281 against newly established BRCA2-deficient mouse mammary tumor cell lines and to determine potential synergy between AZD2281 and cisplatin.

Experimental Design: We established and thoroughly characterized a panel of clonal cell lines from independent BRCA2-deficient mouse mammary tumors and BRCA2-proficient control tumors. Subsequently, we assessed sensitivity of these lines to conventional cytotoxic drugs and the novel PARP inhibitor AZD2281. Finally, in vitro combination studies were done to investigate interaction between AZD2281 and cisplatin.

Results: Genetic, transcriptional, and functional analyses confirmed the successful isolation of BRCA2-deficient and BRCA2-proficient mouse mammary tumor cell lines. Treatment of these cell lines with 11 different anticancer drugs or with γ-irradiation showed that AZD2281, a novel and specific PARP inhibitor, caused the strongest differential growth inhibition of BRCA2-deficient versus BRCA2-proficient mammary tumor cells. Finally, drug combination studies showed synergistic cytotoxicity of AZD2281 and cisplatin against BRCA2-deficient cells but not against BRCA2-proficient control cells.

Conclusion: We have successfully established the first set of BRCA2-deficient mammary tumor cell lines, which form an important addition to the existing preclinical models for BRCA-mutated breast cancer. The exquisite sensitivity of these cells to the PARP inhibitor AZD2281, alone or in combination with cisplatin, provides strong support for AZD2281 as a novel targeted therapeutic against BRCA-deficient cancers.

Mutations in the breast cancer susceptibility genes BRCA1 and BRCA2 are responsible for the majority of hereditary breast cancers (1). Tumors in patients with heterozygous BRCA1 or BRCA2 germ-line mutations typically show somatic loss of heterozygosity at the BRCA1 or BRCA2 locus, respectively, resulting in loss of the wild-type allele (2, 3). Functionally, involvement of BRCA2 in the repair of DNA damage, especially double-strand breaks (DSB), has been firmly established by several groups (46). The primary role of BRCA2 in this process appears to be the regulation of damage-induced RAD51 protein filaments that are required for DSB repair by homologous recombination (7).

In the absence of BRCA2, repair of DSBs by error-prone mechanisms (8) and chromosomal instability due to improper centrosome maintenance (9) result in genomic instability (10), which renders cells susceptible to acquiring additional cancer initiating genetic lesions. The absence of error-free DSB repair mechanisms may prove to be the Achilles heel of BRCA2-deficient tumors, as increased sensitivity to γ-irradiation or DNA-damaging agents is observed in cells with dysfunctional BRCA2 (4, 1113). Clinical trials exploiting the sensitivity of BRCA-mutated breast cancers to platinum-based DNA-damaging drugs have been started recently (14, 15).

Because platinum-based monotherapy is associated with toxicity (16), alternative strategies to stress the DSB repair pathway are urgently needed. An attractive strategy, which has recently gained momentum, is based on inhibition of poly(ADP-ribose) polymerase-1 (PARP-1), which binds to DNA strand breaks and helps regulate base excision repair. On binding, both the protein itself and the surrounding histones are poly ADP-ribosylated, which may help in attracting repair factors, such as the single-strand break repair factor XRCC1 (17, 18). The fact that Parp1-/- mice, in contrast to Xrcc1 mutants, are viable and fertile suggests these mice are not completely deficient in the repair of single-strand breaks (19, 20). In the absence of efficient single-strand break repair, endogenously created lesions may persist through to S phase, where they are converted to DSBs (21). DSB repair is normally intact in PARP1-/- cells (22) but not in BRCA1/2-deficient cells (5, 23). As such, inhibition of PARP-1 may confer selective cytotoxicity to tumor cells with attenuated BRCA function while not affecting on the BRCA-proficient cells of the patient. In support of this, two groups showed selective cytotoxicity of PARP inhibitors against cells with dysfunctional BRCA1/2 (24, 25).

In vitro analysis of PARP inhibitors in CAPAN-1 tumor cells, derived from a liver metastasis of a human BRCA2-defective pancreas carcinoma (26) and thus far the only tumor cell line available with abrogated BRCA2 function, yielded conflicting results (27, 28). These experiments in pancreatic tumor cells call for evaluation of PARP inhibitor efficacy in additional BRCA2-deficient tumor cells, specifically of mammary epithelial origin. The latter is important considering the cell type–intrinsic differences between mammary epithelial cells and other cells, which have been postulated to—at least partially—explain the tumor spectrum in BRCA carriers (29). Unfortunately, no BRCA2-deficient mammary tumor cell lines have been published until now.

Here, we report the successful establishment of clonal cell lines from two independent BRCA2-deficient mouse mammary tumors. Thorough characterization of these cell lines confirmed complete loss of BRCA2 function and increased sensitivity toward DNA-damaging agents was shown for BRCA2-deficient cells compared with BRCA2-proficient control cells. Using a novel, specific, and potent inhibitor of PARP enzymatic activity, AZD2281,4

4

K.A. Menear et al., submitted for publication.

we show growth inhibition of BRCA2-deficient mammary tumor cells. Because both PARP inhibition and cisplatin confer selective toxicity to BRCA2-deficient mammary tumor cells, and because PARP inhibition was recently shown to potentiate cisplatin-mediated cytotoxicity (30), we also did drug combination studies. AZD2281 synergized with cisplatin in inhibiting the growth of BRCA2-deficient mammary tumor cells, whereas this combination was additive in the BRCA2-proficient tumor cells. These data warrant further preclinical evaluation of AZD2281 as monotherapy or in combination with cisplatin in animal models for BRCA-deficient breast cancer.

Establishment and maintenance of tumor cell lines. Tumor-bearing female mice of the K14-Cre;Brca2F11/F11;p53F2-10/F2-10 (KB2P) or K14-Cre;Brca2wt/wt;p53F2-10/F2-10 (KP) genotype (31) were sacrificed and the tumors were isolated. Small (3 × 3 mm) pieces were subsequently minced and digested for 1 h in Leibovitz L15 medium with 3 g/L collagenase A and 1.5 g/L porcine pancreatic trypsin with rigorous shaking at 37°C. Aggregates were plated out and cultured under low oxygen conditions (3% O2, 5% CO2, 37°C) using DMEM/F-12 (Life Technologies) supplemented with 10% FCS, 50 units/mL penicillin, 50 μg/mL streptomycin (Life Technologies), 5 μg/mL insulin (Sigma), 5 ng/mL epidermal growth factor (Life Technologies), and 5 ng/mL cholera toxin (Gentaur). To remove contaminating fibroblasts, cultures were differentially trypsinized until homogeneous cell morphology indicated pure epithelial cultures.

Detection of Brca2 expression by quantitative PCR. Total RNA (1.25 μg) isolated from cell cultures using a Qiagen RNeasy kit was used as input for a first-strand reaction (Invitrogen) according to manufacturer's protocol. Subsequently, 12 ng cDNA was used for a quantitative PCR using the SYBR Green PCR Mastermix (Applied Biosystems) done on an ABI Prism 7000. HPRT levels were used as internal control. Brca2 primer sequences were as follows: exon2-3 forward: gaaatttttaaggcgagatgcag and reverse: ccaattgaggcttatcggtcc, exon10-11 forward: gaagcaagtgcttttgaag and reverse: cagaagaatctggtatacctg, and exon18-19 forward: ctcctgatgcctgtgcacc and reverse: cacgaaagaaccccagcct.

Detection of p53 protein. Protein extraction of cultured cells was done using ELB buffer [150 mmol/L NaCl, 50 mmol/L HEPES (pH 7.5), 5 mmol/L EDTA, 0.1% NP-40] complemented with a protease inhibitor cocktail (Roche). Primary antibodies used in subsequent Western blot assays: polyclonal sheep anti-p53 (1:5,000; Calbiochem), polyclonal goat anti-β-actin (1:2,000; Santa Cruz Biotechnology). Secondary antibody: rabbit anti-goat horseradish peroxidase (1:2,000; DakoCytomation).

γH2A.X/RAD51 colocalization. Cells grown on coverslips were exposed to 20 Gy γ-irradiation and fixed 8 h later using 1% paraformaldehyde in PBS. Cells were permeabilized 5′ in 0.1% Triton and preincubated for 1 h at room temperature in staining buffer (PBS-0.5% bovine serum albumin-0.15% glycine), which was used as solvent in all subsequent steps. Incubation with primary polyclonal rabbit anti-RAD51 antibody (a generous gift by Roland Kanaar, Erasmus Medical Center, Rotterdam, The Netherlands) followed for 2 h. Secondary goat anti-rabbit Alexa 568 (1:400 dilution; Molecular Probes) was then coincubated with FITC-conjugated monoclonal mouse anti-γH2A.X antibody (clone JBW301, 1:50 dilution; Upstate) for 1 h at room temperature. Finally, DNA was stained using 1:5,000 To-Pro-3 (Molecular Probes). All incubations were followed by at least three wash steps using staining buffer. Slides were mounted using Vectashield (Vector Laboratories). Images were acquired on a Leica TCS TNT system (Leica Microsystems).

Array comparative genomic hybridization. Genomic DNA isolated from primary tumors or cell lines was labeled and hybridized to 3K mouse BAC microarrays using spleen DNA originating from the same animal as a reference, as described (32).

Karyotyping. Metaphases were prepared by culturing subconfluent cells for 30 min in the presence of 100 ng/mL colcemid (Life Technologies). Cells were harvested and hypotonic swelling was induced for 10 min in 0.075 mol/L KCl. Subsequently, cells were fixed using three short incubations in 3:1 dry methanol/glacial acetic acid solution and dropped on a microscope slide. Ploidy analysis was done by mounting metaphase slides in 4′,6-diamidino-2-phenylindole-containing Vectashield (Vector Laboratories) and images were acquired using a Zeiss Axiovert 200M fluorescence microscope mounted with a Zeiss Axiocam MRm Rev. 2 camera. Spectral karyotyping analysis was done as described (33) using the spectracube 300. Between six and nine metaphases were analyzed for each cell line using Skyview software version 2.1.1.

Drugs. AZD2281 was synthesized by KuDOS Pharmaceuticals.4 Other compounds were obtained from Mayne Pharma (cisplatin), Sigma-Aldrich (mitomycin C, methylmethane sulfonate, 5-fluorouracil, hydroxyurea, nocodazole, and valproic acid), Schering-Plough (temozolomide), Pharmacia Netherlands (doxorubicin), and Aventis (docetaxel).

Temozolomide was dissolved in 10% ethanol (v/v) in saline to a concentration of 5 mg/mL.

Sulforhodamine B growth inhibition assays. Cells were typically plated out in 96-well microplates on day 0 and either irradiated or supplied with 2-fold serial drug dilutions on day 1. All drugs were left on the cells for the duration of the experiment, except for methylmethane sulfonate, which was removed from the cells after 1-h incubation in serum-free medium. On day 5, the cells were fixed by adding trichloroacetic acid to a final concentration of 5% (v/v). After 1 h at 4°C, plates were washed five times with demi water, dried, and stained for 30 min with 50 μL sulforhodamine B (0.4%, w/v). Following three wash steps with 1% acetic acid, 150 μL Tris (10 mmol/L) was added to dissolve the staining. Absorbance at 540 nm was measured using a Tecan infinite m200 plate reader (Tecan). After correction for medium-only and no-drug controls, data points were fitted using the general formula for a sigmoid curve

\[\mathrm{Survival}=\frac{1}{1+(\frac{[\mathrm{Drug}]}{\mathrm{IC}_{50}})^{m}}\]

using the sigmoidity (m) and IC50 as fit variables and Matlab software (The Mathworks). At least three independent IC50 values were measured for each drug/cell line combination.

Establishment of BRCA2-deficient mammary tumor cell lines.In vitro studies on BRCA2-associated breast cancer have been hampered by the lack of appropriate BRCA2-deficient mammary tumor cell lines (24, 25). We therefore set out to establish BRCA2-deficient mammary tumor cell lines by taking into culture 12 tumors from KB2P mice that develop BRCA2-deficient mammary cancer (31). As controls for functional assays, we also established cell lines from three BRCA2-proficient mammary tumors harvested from KP mice (34). Histopathologic examination of the parental tumors confirmed their resemblance to invasive ductal carcinomas, the main tumor type in human BRCA2 mutation carriers (Supplementary Fig. S1). Low oxygen (3%) culturing conditions were continuously applied to minimize oxidative DNA damage and thereby prevent specific depletion of BRCA2-deficient cells in KB2P cultures (35). When cultures showed homogeneous epithelial characteristics on visual inspection, the KB2P lines were probed for Cre-mediated recombination of the Brca2F11 alleles using Southern analysis (Fig. 1A). Whereas all corresponding primary tumors showed homozygous switching of both Brca2F11 alleles (with a residual Brca2F11 band derived from nonswitched stromal cells), one or two functional Brca2F11 alleles were retained in cell lines established from 10 of 12 tumors, suggesting strong selection against BRCA2-deficient mammary epithelial cells during in vitro culture. Nevertheless, homozygous Brca2 loss was detected in two independent tumor cell lines (KB2P-1 and KB2P-3; Fig. 1A). Limiting dilution culturing resulted in five clonal cell lines with homozygously switched Brca2F11 alleles (Fig. 1B) and three clonal cell lines derived from BRCA2-proficient KP tumors (KP-3.33, KP-6.3, and KP-7.7). The latter served as controls in subsequent experiments. Loss of Brca2 expression in KB2P cell lines was confirmed by quantitative reverse transcription-PCR. In contrast to BRCA2-proficient KP-6.3 cells, both clonal BRCA2-deficient cell lines displayed complete absence of full-length Brca2 expression (Fig. 1C). Interestingly, however, absolute levels of the truncated Brca2Δ11 transcript, which is produced after recombination, appeared to be higher in the KB2P lines, suggesting increased transcription of the Brca2 gene on loss of functional BRCA2 (Fig. 1C). Western blot analysis showed p53 to be absent in all established clonal cell lines, confirming Cre-mediated recombination of both p53F2-10 alleles (Fig. 1D). Epithelial identity of all clonal cell lines was confirmed using cytokeratin-8 and E-cadherin staining of cultured cells (Supplementary Fig. S2).

Fig. 1.

Brca2 and p53 status in cell lines from BRCA2-deficient mammary tumors. A, Southern blot analysis of primary tumors and cell lines for Brca2. Top, genomic structure of the Brca2 locus along with the probe location and BlnI (B)/NheI (N) restriction sites. Triangles, LoxP sites. Expected sizes of bands of all possible alleles detected by Southern blot analysis are indicated. All tumors analyzed showed strong Brca2Δ bands and no or weak Brca2F bands. In contrast, most of the resulting cell lines contained equal amounts of switched/unswitched DNA as expected for heterozygous Brca2F11/Δ11 cells. Only KB2P-1 and KB2P-3 cell lines showed homozygous deletion of Brca2. B, several clones of the KB2P-3 and KB2P-1 lines were reanalyzed using Southern blot analysis and still showed Brca2 loss. C, quantitative PCR analysis of BRCA2-deficient clones KB2P-1.21 and KB2P-3.4 and BRCA2-proficient clonal line KP-6.3 confirm complete loss of full-length Brca2 expression. Three different regions of the Brca2 transcript were analyzed by real-time reverse transcription-PCR using different primer pairs spanning intron/exon boundaries. D, Western blot analysis showed that all clonal KP and KB2P cell lines have lost p53 expression. NIH-3T3 cells were used as a positive control for p53 detection. Immunoblotting for β-actin was used as loading control.

Fig. 1.

Brca2 and p53 status in cell lines from BRCA2-deficient mammary tumors. A, Southern blot analysis of primary tumors and cell lines for Brca2. Top, genomic structure of the Brca2 locus along with the probe location and BlnI (B)/NheI (N) restriction sites. Triangles, LoxP sites. Expected sizes of bands of all possible alleles detected by Southern blot analysis are indicated. All tumors analyzed showed strong Brca2Δ bands and no or weak Brca2F bands. In contrast, most of the resulting cell lines contained equal amounts of switched/unswitched DNA as expected for heterozygous Brca2F11/Δ11 cells. Only KB2P-1 and KB2P-3 cell lines showed homozygous deletion of Brca2. B, several clones of the KB2P-3 and KB2P-1 lines were reanalyzed using Southern blot analysis and still showed Brca2 loss. C, quantitative PCR analysis of BRCA2-deficient clones KB2P-1.21 and KB2P-3.4 and BRCA2-proficient clonal line KP-6.3 confirm complete loss of full-length Brca2 expression. Three different regions of the Brca2 transcript were analyzed by real-time reverse transcription-PCR using different primer pairs spanning intron/exon boundaries. D, Western blot analysis showed that all clonal KP and KB2P cell lines have lost p53 expression. NIH-3T3 cells were used as a positive control for p53 detection. Immunoblotting for β-actin was used as loading control.

Close modal

BRCA2-deficient cell lines do not form irradiation-induced RAD51 foci. Cre-mediated recombination within the Brca2F11 allele results in loss of exon 11, which encodes the BRC repeats necessary for proper regulation of RAD51 protein filament formation during DSB repair by homologous recombination (7, 36). In BRCA2-proficient cells, induction of DSBs by ionizing radiation results in rapid relocalization of both RAD51 and γH2A.X to sites of DNA damage (37). In contrast, relocalization of RAD51 does not occur in irradiated CAPAN-1 cells with dysfunctional BRCA2 (38). To ascertain that Brca2Δ11/Δ11;p53Δ2-10/Δ2-10 KB2P cells were incapable of carrying out homologous recombination–mediated DSB repair, we assessed radiation-induced colocalization of RAD51 with γH2A.X in nuclear foci. BRCA2-proficient cells, subjected to 20 Gy γ-irradiation and harvested 8 hours later, showed profound colocalization of RAD51 and γH2A.X. In contrast, complete absence of irradiation-induced RAD51 foci was seen in the KB2P cells lacking wild-type BRCA2 (Fig. 2).

Fig. 2.

BRCA2-deficient tumor cells show loss of γH2A.X/RAD51 colocalization after irradiation. On irradiation with 20 Gy, BRCA2-proficient KP mammary tumor cells showed colocalization of γH2A.X (green) and RAD51 (red) at sites of DNA damage (blue, DNA). In contrast, BRCA2-deficient KB2P cells show complete absence of RAD51 signals at irradiation-induced γH2A.X nuclear foci.

Fig. 2.

BRCA2-deficient tumor cells show loss of γH2A.X/RAD51 colocalization after irradiation. On irradiation with 20 Gy, BRCA2-proficient KP mammary tumor cells showed colocalization of γH2A.X (green) and RAD51 (red) at sites of DNA damage (blue, DNA). In contrast, BRCA2-deficient KB2P cells show complete absence of RAD51 signals at irradiation-induced γH2A.X nuclear foci.

Close modal

BRCA2-deficient mammary tumor cell lines display genomic instability. Absence of proper DSB repair (8), centrosome regulation (9), and cytokinesis (39) on loss of BRCA2 is thought to result in a mutator phenotype driving additional oncogenic events. Indeed, BRCA2-deficient tumors have been shown to contain high amounts of genetic aberrations (40). To investigate the extent of genomic gains and losses, we did genome-wide array comparative genomic hybridization analysis of both the parental BRCA2 tumors and the clonal cell lines derived thereof (Fig. 3A). Many genomic regions were found to be significantly amplified (red) or deleted (green) compared with control DNA harvested from the spleen of the tumor-bearing animal. In general, the BRCA2-deficient tumors showed much more aberrations than the BRCA2-proficient control tumors, similar to the situation in BRCA1-deficient tumors (34). Unsupervised hierarchical clustering of all parental tumors and KB2P cell lines showed coclustering of the KB2P-1 clones together with their ancestral tumor 1 but not of KB2P-3 clones with tumor 3 (Fig. 3A). For subsequent ploidy analysis, metaphase spreads were prepared and stained with 4′,6-diamidino-2-phenylindole to count at least 18 spreads. All but one cell line (KP-6.3) were shown to be aneuploid, a general hallmark of solid tumors (ref. 41; Fig. 3B). To determine the contribution of individual chromosomes to the increased ploidy as well as the amount of chromosome rearrangements, we did spectral karyotyping analysis. BRCA2-deficient KB2P cells showed large amounts of complex translocations and in addition displayed many other structural aberrations such as deletions and the presence of satellite and marker chromosomes (Fig. 3C and D). Besides the expected increase in structural chromosomal aberrations in the KB2P cells versus the KP cells, we observed more clonal variation within the BRCA2-negative cells than in the controls (Fig. 3C). The complete composite karyotype of several cells analyzed is shown in Supplementary Table S1.

Fig. 3.

Brca2 mutated cell lines display genomic instability. A, array-comparative genomic hybridization analysis was carried out on DNA of primary tumors and clonal cell lines derived from these. Top, log2 ratios of DNA copy number changes compared with spleen control DNA. Red dots, significant copy number gains; green dots, significant copy number losses. Bottom, unsupervised hierarchical clustering of comparative genomic hybridization profiles of all primary KB2P tumors and the clonal cell lines KB2P-1.21 (derived from tumor KB2P-1) and KB2P-3.4 (derived from tumor KB2P-3). Increasing vertical distance indicates larger differences in comparative genomic hybridization profiles. B, ploidy analysis of clonal KP and KB2P cell lines. A minimum of 18 4′,6-diamidino-2-phenylindole-stained metaphases were photographed and chromosomes were counted. Bars, SD. C, spectral karyotyping analysis revealed large amounts of structural and nonclonal aberrations in KB2P cell lines compared with KP controls. Aberrations were considered clonal if more than two metaphases contained a specific aberration or gain or more than three losses were detected. Average numbers of clonal and nonclonal aberrations per cell for both numerical (red) and structural (blue) aberrations. D, a representative spectral karyotyping example of BRCA2-deficient clone KB2P-3.4, showing both numerical and structural chromosomal aberrations.

Fig. 3.

Brca2 mutated cell lines display genomic instability. A, array-comparative genomic hybridization analysis was carried out on DNA of primary tumors and clonal cell lines derived from these. Top, log2 ratios of DNA copy number changes compared with spleen control DNA. Red dots, significant copy number gains; green dots, significant copy number losses. Bottom, unsupervised hierarchical clustering of comparative genomic hybridization profiles of all primary KB2P tumors and the clonal cell lines KB2P-1.21 (derived from tumor KB2P-1) and KB2P-3.4 (derived from tumor KB2P-3). Increasing vertical distance indicates larger differences in comparative genomic hybridization profiles. B, ploidy analysis of clonal KP and KB2P cell lines. A minimum of 18 4′,6-diamidino-2-phenylindole-stained metaphases were photographed and chromosomes were counted. Bars, SD. C, spectral karyotyping analysis revealed large amounts of structural and nonclonal aberrations in KB2P cell lines compared with KP controls. Aberrations were considered clonal if more than two metaphases contained a specific aberration or gain or more than three losses were detected. Average numbers of clonal and nonclonal aberrations per cell for both numerical (red) and structural (blue) aberrations. D, a representative spectral karyotyping example of BRCA2-deficient clone KB2P-3.4, showing both numerical and structural chromosomal aberrations.

Close modal

Selective sensitivity of BRCA2-deficient mammary tumor cell lines to DNA-damaging drugs. Absence of proper repair of damaged DNA leads to hypersensitivity to agents affecting DNA structure. Indeed, the specific hypersensitivity of BRCA2-deficient tumor cells to DNA-damaging drugs is the basis for several clinical trials in BRCA2 mutation carriers with breast or ovarian cancer (14, 15). To evaluate the utility of our mammary tumor cell lines to model BRCA2-deficient breast cancer, sensitivity to both γ-irradiation and cisplatin was determined in growth inhibition assays. Both treatments clearly resulted in strong specific growth inhibition in the BRCA2 mutant KB2P lines compared with the KP controls (Fig. 4).

Fig. 4.

BRCA2 dysfunction strongly sensitizes cells to the effects of AZD2281 and DNA-damaging drugs. Representative growth inhibitory curves for AZD2281, temozolomide, mitomycin C, cisplatin, γ-irradiation, and methylmethane sulfonate. Blue, BRCA2-deficient cells; red, data from BRCA2-proficient cells.

Fig. 4.

BRCA2 dysfunction strongly sensitizes cells to the effects of AZD2281 and DNA-damaging drugs. Representative growth inhibitory curves for AZD2281, temozolomide, mitomycin C, cisplatin, γ-irradiation, and methylmethane sulfonate. Blue, BRCA2-deficient cells; red, data from BRCA2-proficient cells.

Close modal

PARP inhibitor AZD2281 strongly inhibits BRCA2-deficient cell growth. Recent experiments support the idea that inhibiting PARP-1 activity, and thus single-strand break repair, may induce DSBs that are specifically toxic to cells with defective homologous recombination repair, such as BRCA2-deficient cells (24, 25). To test the utility of PARP inhibitors for treating BRCA2-associated breast cancer, we investigated selective growth inhibition of a novel PARP inhibitor, AZD2281, in our panel of BRCA2-proficient and BRCA2-deficient cell lines (Fig. 4; Table 1). To compare the activity of AZD2281 with various classes of cytotoxic agents, we tested a set of 10 drugs, including compounds directly inducing DNA strand lesions that result in DSBs (mitomycin C, methylmethane sulfonate, and temozolomide), antimetabolites (5-fluorouracil and hydroxyurea), spindle poisons (docetaxel and nocodazole), the topoisomerase II inhibitor doxorubicin, and the histone deacetylase inhibitor valproic acid. Analogous to γ-irradiation and cisplatin, the DNA cross-linker mitomycin C and the alkylating agents methylmethane sulfonate and temozolomide resulted in significantly lower IC50 values in KB2P lines compared with KP lines (Fig. 4; Table 1). BRCA2 deficiency did not, on the other hand, significantly potentiate the action of the antimetabolites, the spindle poisons, doxorubicin, and valproic acid. Importantly, the largest difference in drug sensitivity between BRCA2-deficient KB2P cells and BRCA2-proficient KP cells was observed for the PARP inhibitor AZD2281. Data from at least three independent experiments showed an average IC50 of 91 nmol/L for the KB2P lines compared with 8,135 nmol/L for the KP lines, which is a highly significant (P = 2.7 × 10-6) difference of ∼90 times.

Table 1.

Differential IC50 values of 12 agents on Brca2-proficient versus Brca2-deficient cells

AgentBrca2 proficient
Brca2 deficient
RatioSignificance
KP3.33KP6.3KP7.7KB2P3.4KB2P1.21
AZD2281 (nmol/L) 9,453 (1,002) 5,705 (918) 10,428 (559) 57 (19) 124 (35) 89.7* 2.7 × 10−6 
Temozolomide (μmol/L) 1,347 (241) 1,093 (193) ND 12.3 (2.85) 41 (4.75) 45.7* 1.3 × 10−7 
Mitomycin C (nmol/L) 303 (53) 427 (24) 165 (8.7) 25 (1.5) 20 (1.8) 13.3* 1.3 × 10−4 
Cisplatin (nmol/L) 2,124 (382) 1,043 (177) 969 (159) 117 (12) 108 (14) 12.2* 6.1 × 10−4 
Irradiation (Gy) 8.4 (0.64) 5.8 (0.18) 3.9 (0.61) 1.6 (0.14) 1.4 (0.27) 4.1* 0.003 
Methylmethane sulfonate (μmol/L) 759 (7.7) 1,256 (103) 633 (8.5) 196 (13) 343 (54) 3.3* 3.5 × 10−4 
Doxorubicin (nmol/L) 34.7 (3.2) 13.3 (1.8) 5.4 (0.90) 3.3 (0.63) 7.4 (0.53) 0.03 
Docetaxel (nmol/L) 1.53 (0.37) 1.36 (0.10) 0.27 (0.037) 0.23 (0.015) 0.63 (0.13) 2.6 0.03 
Valproic acid (mmol/L) 2.3 (0.17) 2.3 (0.17) 1.6 (0.051) 1.7 (0.40) 1.4 (0.13) 1.3 0.05 
Nocodazole (nmol/L) 55.3 (9.5) 62.0 (6.7) 54.2 (8.9) 49.8 (5.1) 40.9 (5.9) 1.3 0.09 
5-Fluorouracil (μmol/L) 2.3 (0.11) 2.4 (0.067) 2.4 (0.20) 1.9 (0.35) 1.9 (0.15) 1.2 0.02 
Hydroxyurea (μmol/L) 51 (13) 67 (9.9) 29 (3.6) 44 (7.7) 43 (9.5) 1.1 0.58 
AgentBrca2 proficient
Brca2 deficient
RatioSignificance
KP3.33KP6.3KP7.7KB2P3.4KB2P1.21
AZD2281 (nmol/L) 9,453 (1,002) 5,705 (918) 10,428 (559) 57 (19) 124 (35) 89.7* 2.7 × 10−6 
Temozolomide (μmol/L) 1,347 (241) 1,093 (193) ND 12.3 (2.85) 41 (4.75) 45.7* 1.3 × 10−7 
Mitomycin C (nmol/L) 303 (53) 427 (24) 165 (8.7) 25 (1.5) 20 (1.8) 13.3* 1.3 × 10−4 
Cisplatin (nmol/L) 2,124 (382) 1,043 (177) 969 (159) 117 (12) 108 (14) 12.2* 6.1 × 10−4 
Irradiation (Gy) 8.4 (0.64) 5.8 (0.18) 3.9 (0.61) 1.6 (0.14) 1.4 (0.27) 4.1* 0.003 
Methylmethane sulfonate (μmol/L) 759 (7.7) 1,256 (103) 633 (8.5) 196 (13) 343 (54) 3.3* 3.5 × 10−4 
Doxorubicin (nmol/L) 34.7 (3.2) 13.3 (1.8) 5.4 (0.90) 3.3 (0.63) 7.4 (0.53) 0.03 
Docetaxel (nmol/L) 1.53 (0.37) 1.36 (0.10) 0.27 (0.037) 0.23 (0.015) 0.63 (0.13) 2.6 0.03 
Valproic acid (mmol/L) 2.3 (0.17) 2.3 (0.17) 1.6 (0.051) 1.7 (0.40) 1.4 (0.13) 1.3 0.05 
Nocodazole (nmol/L) 55.3 (9.5) 62.0 (6.7) 54.2 (8.9) 49.8 (5.1) 40.9 (5.9) 1.3 0.09 
5-Fluorouracil (μmol/L) 2.3 (0.11) 2.4 (0.067) 2.4 (0.20) 1.9 (0.35) 1.9 (0.15) 1.2 0.02 
Hydroxyurea (μmol/L) 51 (13) 67 (9.9) 29 (3.6) 44 (7.7) 43 (9.5) 1.1 0.58 

NOTE: Mean IC50 values of at least three independent experiments were determined for 11 drugs and γ-irradiation. Values inside brackets are the SE of the independent IC50 determinations. Average IC50 values determined for the KP lines were compared to average IC50 values of the KB2P lines and displayed as their ratio. Statistical significance was determined using a t test.

*

P < 0.01.

For temozolomide, no accurate IC50 values could be obtained for KP-7.7 due to deviation of actual data points from a sigmoid curve.

PARP inhibition synergizes with cisplatin in growth inhibition of BRCA2-deficient mammary tumor cells. Because both platinum drugs and the PARP inhibitor AZD2281 give rise to DSBs, combination therapy with platinum and AZD2281 may lead to a reduced effective dose of both single agents. This can be favorable with regard to toxic side effects of platinum drugs. To study possible drug synergy or additivity, growth inhibition induced by combinations of 16 different concentrations of AZD2281 and 9 different concentrations of cisplatin was determined for both KB2P cell lines and a KP control cell line (Fig. 5A). Subsequently, using the growth inhibition curves of both single agents, combination index values as explained in ref. 42 were calculated using the formula:

Fig. 5.

Cisplatin and AZD2281 show additivity in KP lines and synergism in KB2P lines. A, two-dimensional plots showing growth inhibitory effects exerted by combinations of 15 different AZD2281 concentrations with 9 different cisplatin concentrations. The color scale indicates growth ratios of treated cells compared with mock-treated control cells. B, for all drug combinations tested, log10 values of the combination indices are shown. All data points resulting in <20% or >80% growth are arbitrarily set to 0, because predictions of single-agent concentrations for such effects are intrinsically inaccurate.

Fig. 5.

Cisplatin and AZD2281 show additivity in KP lines and synergism in KB2P lines. A, two-dimensional plots showing growth inhibitory effects exerted by combinations of 15 different AZD2281 concentrations with 9 different cisplatin concentrations. The color scale indicates growth ratios of treated cells compared with mock-treated control cells. B, for all drug combinations tested, log10 values of the combination indices are shown. All data points resulting in <20% or >80% growth are arbitrarily set to 0, because predictions of single-agent concentrations for such effects are intrinsically inaccurate.

Close modal
\[\mathrm{CI}=\frac{\mathrm{Concentration(AZD2281)}}{\sqrt[\mathrm{m(AZD2281)}]{\frac{\mathrm{IC50(AZD2281)}}{\mathrm{survival(AZD2281,Cisplatin)}}\mathrm{{-}IC50(AZD2281)}}}+\frac{\mathrm{Concentration(Cisplatin)}}{\sqrt[\mathrm{m(Cisplatin)}]{\frac{\mathrm{IC50(Cisplatin)}}{\mathrm{survival(AZD2281,Cisplatin)}}\mathrm{{-}IC50(Cisplatin)}}}\]

.

Thus, log10 CI values < -0.15 indicate synergy at any given combination of drug concentration, CI values > 0.15 indicate antagonism, and values between -0.15 and 0.15 represent additive drug interaction (42). Whereas most cisplatin and AZD2281 combinations only exerted additive growth inhibitory effects on the KP-6.3 cell line, synergistic growth inhibition was observed for both BRCA2-deficient cell lines (Fig. 5B).

Tumor-derived cell lines are likely to recapitulate several cell intrinsic properties of tumor cells in vivo (43) and have been used extensively as models for human cancer. For studying human hereditary breast cancer associated with BRCA2 loss-of-function, researchers have thus far relied on the CAPAN-1 line, which carries a 6174delT mutation in one BRCA2 allele accompanied by loss of the wild-type allele (44). This cell line originates from a pancreatic tumor and may therefore be less qualified as a model system for human BRCA2-mutated breast cancer. Growth arrest, probably caused by rapid accumulation of unrepaired DNA damage, makes it very difficult to culture cells with dysfunctional BRCA2 (4). Consequently, no established BRCA2-deficient mammary tumor cell lines have been reported until now. Because supraphysiologic oxygen levels can increase DNA damage (35), we reasoned that cells with dysfunctional BRCA2 should be cultured at low oxygen conditions comparable with the levels present in situ. Although this strategy resulted in the successful establishment of BRCA2-deficient cell lines from two independent primary mammary tumors, cultures derived from 10 additional tumors displayed depletion of BRCA2-deficient cells, suggesting strong selection against BRCA2 dysfunction during in vitro culture. The fact that all mammary tumors and tumor cell lines derived from K14-Cre;Brca2F11/F11;p53F2-10/F2-10 female mice are p53 deficient shows that inactivation of p53-mediated cell cycle checkpoints is not sufficient to permit in vitro growth of BRCA2-deficient mammary tumor cells. Apparently, other factors present in the tumor stroma are required for unhindered proliferation in the absence of functional BRCA2. Endogenous or culture-induced mutations in the successfully established BRCA2-deficient lines may underlie the evasion of stromal dependency. Identification of such mutations might be of therapeutic relevance.

Detailed functional characterization of our newly established BRCA2-deficient mammary tumor cell lines showed that both KB2P cell lines are defective in irradiation-induced RAD51 foci formation. Together with the observed early embryonic lethality of homozygous Brca2Δ11/Δ11 mouse mutants (31), these data indicate that Cre-mediated switching of Brca2F11 results in a nonfunctional allele. In line with this, both BRCA2-deficient KB2P cell lines showed increased genomic instability, characterized by large numbers of clonal structural aberrations, compared with BRCA2-proficient KP cell lines. The high degree of genomic instability of the BRCA2-deficient KB2P cell lines is also reflected by the fact that many aberrations were not clonal, giving rise to de novo genetic heterogeneity within the clonal KB2P cell lines. This continuing genetic heterogeneity may, in combination with strong selective pressure for adaptation to in vitro growth conditions, also be the cause of the discordance between the comparative genomic hybridization profiles of KB2P-3 clonal cell lines and the corresponding primary tumor.

Interestingly, whereas triradial and quadriradial chromosome structures have been described to be a direct effect of BRCA2 dysfunction (4), these structures were not observed in any of the metaphase spreads prepared from our KB2P cell lines (data not shown).

Although it is known that DNA-damaging agents show selective cytotoxicity against nontumor cells with engineered Brca2 mutations, relatively few data are available about the effect of such agents on BRCA2-deficient mammary tumor cells. These tumor cells are either intrinsically less sensitive to the proliferative impediment that is induced by BRCA2 loss-of-function in nontumor cells or have somehow overcome this inhibition perhaps by acquiring additional mutations. For this reason, it is important to assess the efficacy of known cytotoxic agents in the context of steadily proliferating BRCA2-deficient cells. We have assessed, for the first time, the selective toxicity of γ-irradiation and 11 anticancer drugs on proliferating BRCA2-deficient breast cancer cells. Our in vitro cytotoxicity studies with BRCA2-proficient KP cell lines versus BRCA2-deficient KB2P lines clearly showed selective sensitivity of KB2P cells to cytotoxic agents directly inducing DNA strand lesions (irradiation, cisplatin, mitomycin C, methylmethane sulfonate, and temozolomide) but not to agents that do not, or only indirectly, induce DNA damage (doxorubicin, 5-fluorouracil, hydroxyurea, docetaxel, nocodazole, and valproic acid). Temozolomide displayed an exquisite 46-fold higher sensitivity toward BRCA2-deficient cells. This strong selectivity may be explained by the fact that temozolomide indirectly induces DSBs as a result of futile mismatch repair of O6-methylguanine lesions (45). Our results highlight, for the first time, the potential efficacy of temozolomide against BRCA-mutated breast cancer and warrant further investigation. Still, the strongest selective sensitivity of KB2P cells was observed with the clinical PARP inhibitor AZD2281. One explanation for the unrivaled sensitivity of BRCA2-defective cells toward AZD2281 is that suppression of base excision repair by PARP inhibition may result in the conversion of single-strand breaks to DSBs during DNA replication, thus activating BRCA2-dependent recombination pathways (21). The two independently isolated BRCA2-deficient cell lines show very similar responses to all agents tested, as do the three independent BRCA2-proficient lines. Nevertheless, definitive proof of BRCA2 directly being responsible for the differential effects between these groups of cell lines should come from reconstitution experiments that should rule out that other mutations are causal to the observed sensitization spectrum.

The initial reports on the selective sensitivity of BRCA2-deficient cells to PARP inhibition were challenged by other studies, which claimed that known PARP inhibitors were not or only minimally selective to BRCA2 mutant CAPAN-1 tumor cells and BRCA1 mutant tumor cells (27, 46). Although CAPAN-1 cells were subsequently shown to be highly sensitive to the PARP inhibitor used in one of the original studies (28), these experiments did not rule out the possibility that cell type–related differences might compromise the utility of PARP inhibitors for treating BRCA-associated breast cancer. This concern is alleviated by the potent growth inhibition induced by the clinical PARP inhibitor AZD2281 in our BRCA2-deficient mammary cancer cell lines. Whether the strong growth inhibition induced by AZD2281 also applies to BRCA1-deficient tumor cells remains yet to be determined.

Most known PARP inhibitors inhibit PARP-1 as well as PARP-2 activity (24), and in contrast to PARP-1 single knockouts, PARP-1/PARP-2 double-knockout mice are embryonic lethal (47). Indeed, PARP-1 and PARP-2 have both overlapping and nonredundant functions in the maintenance of genomic stability (48). Nevertheless, both the large differential effect between BRCA2-deficient and BRCA2-proficient cells and the absence of apparent toxic effects in mice treated with AZD2281 (data not shown) indicate that partial inhibition of PARP-1 and PARP-2 by AZD2281 does not cause notable toxicity in wild-type cells.

In summary, we have generated the first set of BRCA2-deficient mammary tumor cell lines, which may prove useful for mechanistic studies and for preclinical evaluation of novel anticancer drugs. In vitro cytotoxicity studies with these cell lines showed clear selective sensitivity of BRCA2-deficient cells to drugs that induce DNA strand lesions or compounds that target the base excision repair pathway. Moreover, potent synergy between the clinical PARP inhibitor AZD2281 and cisplatin was observed. Taken together, our results provide strong evidence for the use of selective PARP inhibitors alone or in combination with platinum drugs for the treatment of BRCA-associated cancers and BRCA-like tumors with defective homologous recombination repair.

G.C.M. Smith, N.M.B. Martin, A. Lau and M.J. O'Connor are employees of KuDOS Pharmaceuticals, a wholly owned subsidiary of AstraZeneca.

Grant support: Netherlands Organization for Scientific Research Vidi grant 917.036.347, Dutch Cancer Society grants NKI 2002-2635 and NKI 2007-3772, and Susan G. Komen for the Cure grant BCTR 0403230.

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.

Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).

Current address for P.W.B. Derksen: Department of Medical Oncology, Laboratory of Experimental Oncology, UMC Utrecht, Utrecht, The Netherlands.

We thank Sheba Agarwal and Roland Kanaar for the RAD51 antibody; Liesbeth van Deemter for expert help with generating the KP-3.33, KP-6.3, and KP-7.7 cell lines; and Karin de Visser, Gilles Doumont, Marieke Vollebergh, Jelle Wesseling, Piet Borst, and Hein te Riele for helpful comments on the manuscript.

1
Ford D, Easton DF, Stratton M, et al. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. The Breast Cancer Linkage Consortium.
Am J Hum Genet
1998
;
62
:
676
–89.
2
Collins N, McManus R, Wooster R, et al. Consistent loss of the wild type allele in breast cancers from a family linked to the BRCA2 gene on chromosome 13q12-13.
Oncogene
1995
;
10
:
1673
–5.
3
Smith SA, Easton DF, Evans DG, Ponder BA. Allele losses in the region 17q12-21 in familial breast and ovarian cancer involve the wild-type chromosome.
Nat Genet
1992
;
2
:
128
–31.
4
Patel KJ, Yu VP, Lee H, et al. Involvement of Brca2 in DNA repair.
Mol Cell
1998
;
1
:
347
–57.
5
Moynahan ME, Pierce AJ, Jasin M. BRCA2 is required for homology-directed repair of chromosomal breaks.
Mol Cell
2001
;
7
:
263
–72.
6
Xia F, Taghian DG, DeFrank JS, et al. Deficiency of human BRCA2 leads to impaired homologous recombination but maintains normal nonhomologous end joining.
Proc Natl Acad Sci U S A
2001
;
98
:
8644
–9.
7
Davies AA, Masson JY, McIlwraith MJ, et al. Role of BRCA2 in control of the RAD51 recombination and DNA repair protein.
Mol Cell
2001
;
7
:
273
–82.
8
Tutt A, Bertwistle D, Valentine J, et al. Mutation in Brca2 stimulates error-prone homology-directed repair of DNA double-strand breaks occurring between repeated sequences.
EMBO J
2001
;
20
:
4704
–16.
9
Tutt A, Gabriel A, Bertwistle D, et al. Absence of Brca2 causes genome instability by chromosome breakage and loss associated with centrosome amplification.
Curr Biol
1999
;
9
:
1107
–10.
10
Tutt AN, van Oostrom CT, Ross GM, van Steeg H, Ashworth A. Disruption of Brca2 increases the spontaneous mutation rate in vivo: synergism with ionizing radiation.
EMBO Rep
2002
;
3
:
255
–60.
11
Sharan SK, Morimatsu M, Albrecht U, et al. Embryonic lethality and radiation hypersensitivity mediated by Rad51 in mice lacking Brca2.
Nature
1997
;
386
:
804
–10.
12
Donoho G, Brenneman MA, Cui TX, et al. Deletion of Brca2 exon 27 causes hypersensitivity to DNA crosslinks, chromosomal instability, and reduced life span in mice.
Genes Chromosomes Cancer
2003
;
36
:
317
–31.
13
Bartz SR, Zhang Z, Burchard J, et al. Small interfering RNA screens reveal enhanced cisplatin cytotoxicity in tumor cells having both BRCA network and TP53 disruptions.
Mol Cell Biol
2006
;
26
:
9377
–86.
14
Garber JE, Richardson A, Harris LN, et al. Neo-adjuvant cisplatin (CDDP) in “triple-negative” breast cancer (BC).
Breast Cancer Res Treat
2007
;
100
:
S149
.
15
Yap TA, Boss DS, Fong PC, et al. First in human phase I pharmacokinetic (PK) and pharmacodynamic (PD) study of KU-0059436 (Ku), a small molecule inhibitor of poly ADP-ribose polymerase (PARP) in cancer patients (p), including BRCA1/2 mutation carriers.
J Clin Oncol ASCO Annu Meet Proc Part I
2007
;
25
:
3529
.
16
Kelland L. The resurgence of platinum-based cancer chemotherapy.
Nat Rev Cancer
2007
;
7
:
573
–84.
17
Helleday T, Bryant HE, Schultz N. Poly(ADP-ribose) polymerase (PARP-1) in homologous recombination and as a target for cancer therapy.
Cell Cycle
2005
;
4
:
1176
–8.
18
El Khamisy SF, Masutani M, Suzuki H, Caldecott KW. A requirement for PARP-1 for the assembly or stability of XRCC1 nuclear foci at sites of oxidative DNA damage.
Nucleic Acids Res
2003
;
31
:
5526
–33.
19
Tebbs RS, Flannery ML, Meneses JJ, et al. Requirement for the Xrcc1 DNA base excision repair gene during early mouse development.
Dev Biol
1999
;
208
:
513
–29.
20
Wang ZQ, Auer B, Stingl L, et al. Mice lacking ADPRT and poly(ADP-ribosyl)ation develop normally but are susceptible to skin disease.
Genes Dev
1995
;
9
:
509
–20.
21
Arnaudeau C, Lundin C, Helleday T. DNA double-strand breaks associated with replication forks are predominantly repaired by homologous recombination involving an exchange mechanism in mammalian cells.
J Mol Biol
2001
;
307
:
1235
–45.
22
Yang YG, Cortes U, Patnaik S, Jasin M, Wang ZQ. Ablation of PARP-1 does not interfere with the repair of DNA double-strand breaks, but compromises the reactivation of stalled replication forks.
Oncogene
2004
;
23
:
3872
–82.
23
Moynahan ME, Chiu JW, Koller BH, Jasin M. Brca1 controls homology-directed DNA repair.
Mol Cell
1999
;
4
:
511
–8.
24
Bryant HE, Schultz N, Thomas HD, et al. Specific killing of BRCA2-deficient tumours with inhibitors of poly(ADP-ribose) polymerase.
Nature
2005
;
434
:
913
–7.
25
Farmer H, McCabe N, Lord CJ, et al. Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy.
Nature
2005
;
434
:
917
–21.
26
Fogh J, Fogh JM, Orfeo T. One hundred and twenty-seven cultured human tumor cell lines producing tumors in nude mice.
J Natl Cancer Inst
1977
;
59
:
221
–6.
27
Gallmeier E, Kern SE. Absence of specific cell killing of the BRCA2-deficient human cancer cell line CAPAN1 by poly(ADP-ribose) polymerase inhibition.
Cancer Biol Ther
2005
;
4
:
703
–6.
28
McCabe N, Lord CJ, Tutt AN, Martin NM, Smith GC, Ashworth A. BRCA2-deficient CAPAN-1 cells are extremely sensitive to the inhibition of poly (ADP-ribose) polymerase: an issue of potency.
Cancer Biol Ther
2005
;
4
:
934
–6.
29
Evers B, Jonkers J. Mouse models of BRCA1 and BRCA2 deficiency: past lessons, current understanding and future prospects.
Oncogene
2006
;
25
:
5885
–97.
30
Donawho CK, Luo Y, Luo Y, 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
–37.
31
Jonkers J, Meuwissen R, van der Gulden H, Peterse H, van der Valk M, Berns A. Synergistic tumor suppressor activity of BRCA2 and p53 in a conditional mouse model for breast cancer.
Nat Genet
2001
;
29
:
418
–25.
32
Chung YJ, Jonkers J, Kitson H, et al. A whole-genome mouse BAC microarray with 1-Mb resolution for analysis of DNA copy number changes by array comparative genomic hybridization.
Genome Res
2004
;
14
:
188
–96.
33
Loonstra A, Vooijs M, Beverloo HB, et al. Growth inhibition and DNA damage induced by Cre recombinase in mammalian cells.
Proc Natl Acad Sci U S A
2001
;
98
:
9209
–14.
34
Liu X, Holstege H, van der Gulden H, et al. Somatic loss of BRCA1 and p53 in mice induces mammary tumors with features of human BRCA1-mutated basal-like breast cancer.
Proc Natl Acad Sci U S A
2007
;
104
:
12111
–6.
35
Halliwell B, Aruoma OI. DNA damage by oxygen-derived species. Its mechanism and measurement in mammalian systems.
FEBS Lett
1991
;
281
:
9
–19.
36
Shivji MK, Davies OR, Savill JM, Bates DL, Pellegrini L, Venkitaraman AR. A region of human BRCA2 containing multiple BRC repeats promotes RAD51-mediated strand exchange.
Nucleic Acids Res
2006
;
34
:
4000
–11.
37
Paull TT, Rogakou EP, Yamazaki V, Kirchgessner CU, Gellert M, Bonner WM. A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage.
Curr Biol
2000
;
10
:
886
–95.
38
Yuan SS, Lee SY, Chen G, Song M, Tomlinson GE, Lee EY. BRCA2 is required for ionizing radiation-induced assembly of Rad51 complex in vivo.
Cancer Res
1999
;
59
:
3547
–51.
39
Daniels MJ, Wang Y, Lee M, Venkitaraman AR. Abnormal cytokinesis in cells deficient in the breast cancer susceptibility protein BRCA2.
Science
2004
;
306
:
876
–9.
40
Tirkkonen M, Johannsson O, Agnarsson BA, et al. Distinct somatic genetic changes associated with tumor progression in carriers of BRCA1 and BRCA2 germ-line mutations.
Cancer Res
1997
;
57
:
1222
–7.
41
Rajagopalan H, Lengauer C. Aneuploidy and cancer.
Nature
2004
;
432
:
338
–41.
42
Chou TC. Theoretical basis, experimental design, and computerized simulation of synergism and antagonism in drug combination studies.
Pharmacol Rev
2006
;
58
:
621
–81.
43
Lacroix M, Leclercq G. Relevance of breast cancer cell lines as models for breast tumours: an update.
Breast Cancer Res Treat
2004
;
83
:
249
–89.
44
Goggins M, Schutte M, Lu J, et al. Germline BRCA2 gene mutations in patients with apparently sporadic pancreatic carcinomas.
Cancer Res
1996
;
56
:
5360
–4.
45
Drablos F, Feyzi E, Aas PA, et al. Alkylation damage in DNA and RNA-repair mechanisms and medical significance.
DNA Repair (Amst)
2004
;
3
:
1389
–407.
46
De Soto JA, Wang X, Tominaga Y, et al. The inhibition and treatment of breast cancer with poly (ADP-ribose) polymerase (PARP-1) inhibitors.
Int J Biol Sci
2006
;
2
:
179
–85.
47
Menissier de Murcia J, Ricoul M, Tartier L, et al. Functional interaction between PARP-1 and PARP-2 in chromosome stability and embryonic development in mouse.
EMBO J
2003
;
22
:
2255
–63.
48
Huber A, Bai P, de Murcia JM, de Murcia G. PARP-1, PARP-2 and ATM in the DNA damage response: functional synergy in mouse development.
DNA Repair (Amst)
2004
;
3
:
1103
–8.