The FASEB Journal Express article fj.06-5916fje. Published online June 29, 2006.

The FASEB Journal FJ Express Full-Length Article

Inhibition of poly(ADP-ribose) polymerase prevents irinotecan-induced intestinal damage and enhances irinotecan/temozolomide efcacy against colon carcinoma
Lucio Tentori,* Carlo Leonetti, Marco Scarsella, Alessia Muzi,* Emanuela Mazzon, Matteo Vergati,* Olindo Forini,* Rena Lapidus, Weizheng Xu, Annalisa Susanna Dorio,* Jie Zhang, Salvatore Cuzzocrea, and Grazia Graziani*,1
*Department of Neuroscience, University of Rome Tor Vergata, Rome, Italy; Experimental Clinical Laboratory, Institute for Cancer Research Regina Elena, Rome, Italy; Department of Clinical and Experimental Medicine and Pharmacology, University of Messina and Centro Neurolesi Bonino-Pulejo (IRCCS), Messina, Italy; and MGI Pharma, Baltimore, Maryland, USA
ABSTRACT Poly(ADP-ribose) polymerase (PARP) inhibitors enhance the antitumor activity of the topoisomerase I inhibitor irinotecan (CPT-11), which is used to treat advanced colorectal carcinoma. Since PARP inhibitors sensitize tumor cells also to the methylating agent temozolomide (TMZ) and clinical trials are evaluating CPT-11 in combination with TMZ, we tested whether the PARP inhibitor GPI 15427 (10-(4-methyl-piperazin-1-ylmethyl)-2H-7-oxa-1,2-diaza-benzo[de]anthracen-3-one) increases the efcacy of CPT-11 TMZ against colon cancer. Moreover, due to the ability of PARP inhibitors to avoid cell death consequent to PARP-1 overactivation, we evaluated whether oral administration of GPI 15427 provides protection from the dose-limiting intestinal toxicity of CPT-11. The results of colony formation assay indicated that GPI 15427 increased the antiproliferative effects (combination index <1) of TMZ SN-38 (the active metabolite of CPT-11) against colon cancer cells. Accordingly, GPI 15427 (40 mg/kg/day 5 days per os) in combination with TMZ (10 mg/kg/day 5 days) CPT-11 (4 mg/kg/day 5 days) signicantly reduced the growth of tumor xenografts. Oral administration of GPI 15427 (40 mg/kg/q2 3 days) prevented intestinal injury and diarrhea induced by CPT-11 (30 mg/kg/day 3 days) reducing inammation and PARP-1 overactivation, as evidenced by immunohistochemical staining of intestinal tissue with antipoly(ADP-ribose) antibody (Ab). In conclusion, the PARP inhibitor represents a novel strategy to enhance the antitumor efcacy and reduce toxicity of chemotherapy in colon cancer.Tentori, L., Leonetti, C., Scarsella, M., Muzi, A., Mazzon, E., Vergati, M., Forini, O., Lapidus, R., Xu, W., Dorio, A. S., Zhang, J., Cuzzocrea, S., Graziani, G. Inhibition of poly(ADP-ribose) polymerase prevents irinotecan-induced intestinal damage and enhances irinotecan/temozolomide efcacy against colon carcinoma. FASEB J. 20, E000 E000 (2006)

Colorectal cancer is one of the most common gastrointestinal tract malignancies, and irinotecan (CPT-11) is a cornerstone drug in management of the advanced disease. In fact, CPT-11 in combination with 5-uorouracil/leucovorin is now considered the standard chemotherapy for metastatic colorectal cancer. Moreover, CPT-11 induces remission in 1530% of tumors resistant to 5-uorouracil (1). CPT-11 is a semisynthetic water-soluble derivative of the natural alkaloid camptothecin that requires activation by a carboxylesterase to yield 7-ethyl-10-hydroxycamptothecin (SN-38), a potent topoisomerase I inhibitor. Topoisomerase I cleaves a single strand of DNA and forms a short-lived intermediate with the nick (cleavable complex), allowing rotation of DNA around the noncleaved strand and relaxation of the torsional strain. Religation of strand break restores DNA integrity and is followed by enzyme dissociation from the double helix. Inhibitors of topoisomerase I noncovalently bind to topoisomerase I-DNA, stabilize the cleavable complex and inhibit the religation step. Since this interaction is reversible, cytotoxicity occurs only when the replication fork encounters a cleavable complex, which is converted into a permanent double strand break. Even though CPT-11 has shown activity also against a number of refractory tumors, resistance to this compound remains a critical problem (2). A recent approach to counteract resistance, which is currently evaluated in clinical trials, relies on CPT-11 combination with the anticancer agent temozolomide (TMZ), due to synergistic effects observed in preclinal models (37). TMZ is a DNA methylating agent that generates a wide spectrum of base adducts mainly represented by
Correspondence: Department of Neuroscience, University of Rome Tor Vergata, Via Montpellier 1, Rome 00133, Italy. E-mail: graziani@uniroma2.it doi: 10.1096/fj.06-5916fje
E1
1

Key Words: chemotherapy cancer diarrhea mucositis

0892-6638/06/0020-0001 FASEB

N7-methylguanine, N3-methyladenine, and O6-methylguanine (8). Despite being produced in low amounts, O6-methylguanine is generally considered the main cytotoxic lesion produced by TMZ. In fact, resistance to TMZ is due to high levels of O6-alkylguanine DNA alkyltransferase, which transfers the alkyl adduct from the O6 position of guanine to an internal cysteine residue, or due to functional defects of the mismatch repair system (MR) (9). If not repaired by O6-alkylguanine DNA alkyltransferase, O6-methylguanine inappropriately pairs with thymine and the resultant mismatches trigger the intervention of MR, which removes the improperly paired pyrimidine to reinsert it again opposite to O6-methylguanine. Repeated cycles of MRmediated excision/resynthesis eventually provoke DNA nicks with apoptosis induction and growth arrest. MR deciency has instead been associated with increased sensitivity to CPT-11, probably due to the involvement of MR components in the recombinational repair of DNA double strand breaks generated by the topoisomerase I inhibitor (10, 11). A novel molecular approach to enhance the antitumor activity of topoisomerase I inhibitors or methylating agents relies on the use of chemical inhibitors of poly(ADP-ribose) polymerase (PARP) (1216). PARP denes a family of enzymes that cleaves NAD to nicotinamide and ADP-ribose to form long and branched (ADP-ribose) polymers on glutamic acid residues of a number of proteins. Poly (ADP-ribosyl)ation is involved in the regulation of many cellular processes such as DNA repair, gene transcription, cell cycle progression, cell death, chromatin functions, and genomic stability. In the case of TMZ, the enhancing effect deriving from inhibition of poly(ADP-ribosyl)ation derives from impairment of the repair of Nmethylpurines that generally do not contribute to cytotoxicity. In fact, these methyl adducts are promptly removed by the short-patch base excision repair (BER), which requires PARP-1 and PARP-2 function. During the repair process, PARP-1 and PARP-2 intervention takes place soon after the action of the N-methylpurine DNA glycosylase and apurinic/apyrimidinic endonuclease and is triggered by the DNA nicks resulting from the interruption of sugar-phosphate backbone. Replacement of the damaged nucleotide occurs by means of the coordinate intervention of PARP-1, PARP-2, DNA polymerase , XRCC1, and ligase III (17). Inhibition of PARP-1 and PARP-2 prevents completion of the repair
TABLE 1.
Cell line

process with generation of permanent single strand breaks. The molecular mechanisms underlying tumor chemosensitization to topoisomerase I poisons by PARP inhibitors have been in part claried by recent ndings showing that PARP-1 interacts with and promotes the activity of topoisomerase I and that poly(ADP-ribos)ylated poly (ADP-ribosyl)ated PARP-1 and PARP-2 counteract camptothecin action facilitating resealing of DNA strand breaks (18, 19). Consequently, PARP inhibition hampers topoisomerase I activity favoring the action of the enzyme poisons. A major concern deriving from the use of biomodulators of resistance is that they can potentially increase toxicity exerted by chemotherapy toward normal tissues. However, this might not be the case for PARP inhibitors since these compounds have been shown to protect normal tissues from untoward effects induced by certain genotoxic agents that induce PARP-1 overactivation (20). In fact, elevated DNA damage, which saturates cell repair ability, triggers PARP-1 overactivation with consequent extensive NAD consumption during the synthesis of (ADP-ribose) polymers, which leads to ATP depletion and cell death (21). The aim of the present study was to establish whether the orally bioavailable PARP inhibitor GPI 15427 (10-(4-methyl-piperazin-1ylmethyl)-2H-7-oxa-1,2-diaza-benzo[de]anthracen-3-one, MGI Pharma, Baltimore, MD, USA) (13, 16) enhances the antitumor efcacy of CPT-11 and TMZ combination against colon cancer and provides protection from the intestinal dose-limiting toxicity of CPT-11.

MATERIALS AND METHODS

Cell lines Colon cancer cell lines and their MR functional status are indicated in Table 1. Cultures were maintained in Dulbeccos modied Eagle medium (DMEM) (Sigma-Aldrich, Milan, Italy) supplemented with 10% fetal calf serum, 2 mM Lglutamine, and antibiotics. O6-alkylguanine DNA alkyltransferase activity assay Cells (1 106) were lysed in 0.5 ml of a buffer containing 0.5% 3-[(3-cholamidopropyl) dimethylammonio]propanesulfonate, 50 mM Tris-HCl pH 8, 1 mM EDTA, 3 mM dithiothreitol, 100 mM NaCl, 10% glycerol, protease inhibitors and incu-

Mismatch repair functional status of colon cancer cell lines

Mismatch repair status Reference

LoVoa HT-29a HCT-15a HCT-116b HCT-116 Chr3b,c

a

Decient (homozygous deletion in hMSH2) Procient Decient (mutated hMSH-6) Decient (mutated hMLH1) Procient

44 45 46 47 48

LoVo, HT-29, and HCT-15 cell lines were obtained from ATCC-LGC Promochem (Milan, Italy). b HCT-116 and HCT-116 Chr3 were kindly provided by Prof. Josef Jiricny (University of Zrich, Switzerland). cHCT116 Chr3 was created by microcell chromosome transfer of a single normal human chromosome 3 into HCT-116.

E2

Vol. 20

August 2006

The FASEB Journal

TENTORI ET AL.

bated at 4C for 30 min (22). Various amounts of cell extracts were incubated with 10 g of calf thymus DNA previously labeled with N-[3H]-methyl-N-nitrosourea (GE Healthcare, Milan, Italy; 18 Ci/mmol). O6-alkylguanine DNA alkyltransferase activity was determined by measuring the transfer of [3H]-methyl groups from methylated DNA to O6-alkylguanine DNA alkyltransferase and expressed as fmol of methyl groups per mg of proteins in cell extract. N-methylpurine DNA glycosylase activity assay Cells (1 107) were sonicated at 4C in 0.1 ml buffer I (50 mM Tris-HCl, 3 mM dithiothreitol, and 2 mM EDTA, pH 8.3), with freshly added 1 mM 4-(2-aminoethyl)-benzene-sulfonyl uoride hydrochloride. Various amounts of cell extracts were incubated with 15 g of freshly dissolved calf thymus DNA methylated by N-[3H]methyl-N-nitrosourea in a total volume of 100 l of buffer II (20 mM Tris-HCl, 1 mM dithiothreitol, 60 mM NaCl, and 1 mM EDTA, pH 8). N-methylpurine DNA glycosylase activity was expressed as fmol of methylpurines released per mg of proteins. PARP activity assay Cells (5 106) were lysed in 0.5 ml of a buffer containing 0.1% Triton X, 50 mM Tris-HCl pH 8, 0.6 mM EDTA, 14 mM -mercaptoethanol, 10 mM MgCl2, and protease inhibitors. Proteins (25 g) were incubated with 2 Ci 32P-NAD (GE Healthcare), 10 M NAD , 50 mM Tris-HCl, 10 mM MgCl2, 14 mM -mercaptoethanol, 10 g nuclease-treated salmon testes DNA. PARP activity was expressed as fmol of 32PNAD / g of protein (13). For analysis of GPI 15427 on cellular PARP, intact cells (5 105 tumor cells) were treated with the inhibitor and permeabilized with digitonin (0.1 mg/ml) in the presence of 0.25 Ci 3H-NAD (PerkinElmer, Milan, Italy) (23). For in vivo PARP inhibition, peripheral blood lymphocytes PBL, from untreated or GPI 15427 (40 mg/kg per os)-treated mice (8/group), were separated from whole blood by FicollHypaque density gradient, permeabilized and incubated with 3 H-NAD in the presence of 5 g palindromic deoxyoligonucleotide (CGGAATTCCG). Northern and Western blot analysis Northern blot analysis was performed using total RNA (15 g) and cDNA probes corresponding to the DNA binding domain of PARP-1 (kindly provided by Prof. Alexander Burkle, University of Konstanz, Germany) or to glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Western blot analysis was performed with monoclonal antibodies directed against PARP-1 (BD Biosciences, Milan, Italy), breast cancer resistance protein (Chemicon, Temecula, CA, USA), and actin (Sigma-Aldrich). Signals were quantied using a Kodak densitometer (Rochester, NY, USA). Colony formation assay Cells were seeded in triplicate into 6-well plates (2 102/well) and, after overnight incubation, treated with GPI 15427 (2 M), TMZ (Schering-Plough, Kenilworth, NJ, USA; 311000 M) or SN-38 (Alexis, Florence, Italy; 0.15 nM). When used in combination with TMZ, SN-38 was added to cultures 1 h after TMZ exposure. For experiments to assess the ability of GPI 15427 to enhance the antiproliferative effects of TMZ or SN-38, the PARP inhibitor was added to cell cultures 15 min before the anticancer drugs. Cells were cultured to allow
PARP INHIBITOR FOR CHEMOSENSITIZATION/PROTECTION

colony formation, and after 10 14 days colonies were xed and stained with rhodamine B basic violet 10 (Sigma-Aldrich). Only colonies comprising 50 cells were scored as survival colonies. Chemosensitivity was evaluated in terms of IC50, i.e., the concentration of the drug capable of inhibiting colony-forming ability by 50%. To evaluate whether the combination TMZ SN-38 or TMZ SN-38 GPI 15427 was synergic, tumor cells were exposed to TMZ or SN-38 alone or in combination at xed equipotent ratios (corresponding to 1, 0.5, 0.2, 0.1 the IC50 for each drug) in the absence or presence of a nontoxic dose of GPI 15427 (2 M). The dose-effect curves were analyzed by the median-effect method of Chou and Talalay using the CalcuSyn Software (Biosoft, Cambridge, UK). The combination index (CI) indicates a quantitative measure of the degree of drug interaction in terms of synergistic (CI 1), additive (CI 1), or antagonistic effect (CI 1). Graphs were generated plotting CI as a function of the fraction of cells affected (Fa) by the dose of the anticancer drugs. In vivo studies of antitumor activity HT-29 or LoVo cells (1 106) were inoculated intramuscle (i.m.) in male athymic cluster of differentiation CD-1 mice (nu/nu genotype, 8/group). Tumors were measured with caliper and treatment started when nodules reached 200 400 mm3. Xenograft growth was monitored by measuring tumor nodules every 2 4 days for 3 wk. TMZ was dissolved in dimethyl-sulfoxide, diluted in saline (0.5 mg/ml) and administered intraperitoneally (i.p.) at 10 mg/kg/day 5 days and CPT-11 (Campto, Aventis, Milan, Italy) i.p at 4 mg/kg/day 5 days, doses corresponding to 1/15 of the maximum tolerated dose of TMZ GPI 15427 per os and 1/10 of the maximum tolerated dose of CPT-11 GPI 15427 per os. GPI 15427 was dissolved in 70 mM PBS without potassium and administered by oral gavage once a day (40 mg/kg) for 5 days, 1 h before TMZ or CPT-11. Since the synergistic effect deriving from TMZ and CPT-11 combination is schedule dependent (4), CPT-11 was administered 1 h after TMZ. Treatment was repeated for two cycles. Control mice were always treated with vehicles. Toxicity was assessed on the basis of apparent drug-related deaths (within 7 days after the last treatment) and net body wt loss [i.e., (initial wt-lowest wt)/initial wt 100%]. All procedures involving animal care were performed in compliance with national and international guidelines (European Economy Community Council Directive 86/109, OLJ318, Dec. 1, 1987). Myelosuppression assay Athymic nu/nu mice (6/group) received vehicle, TMZ (100 mg/kg/day 5 days), GPI 15427 (40 mg/kg/day 5 days) or GPI 15427 TMZ. Three mice per group were sacriced on day 8 or 15 and whole blood was analyzed for complete blood count. Intestinal toxicity assay Wistar rats (10/group) were treated i.p. with CPT-11 (30 mg/kg/day 3 days) (24) and received vehicle or two oral doses of GPI 15427 (40 mg/kg) 30 min before and 1 h after each CPT-11 treatment. The day after the last treatment, rats were anesthetized with urethane (1.4 g/kg/i.p.) and jejunum was collected. A 0.5 cm segment of intestine was xed in 10% buffered formalin and embedded into parafn wax. Serial 7 m sections were mounted on silane-coated glass slides. Sections were dewaxed in xylene, rehydrated in graded
E3

ethanol, stained with hematoxylin/eosin, and analyzed using light microscopy (Zeiss, Milan Italy). For analysis of (ADP-ribose) polymers by immunohistochemistry, sections were dewaxed, permeabilized with 0.1% (w/v) Triton X-100 in PBS for 20 min, and incubated overnight with an anti-(ADP-ribose) polymers Ab (Alexis, 1:50 in PBS, v/v). Specic labeling was detected with a biotinconjugated goat anti-rabbit IgG and avidin-biotin peroxidase complex (DBA, Milan, Italy). Densitometry of (ADP-ribose) polymers staining was performed as described previously using an Imaging Densitometer (AxioVision, Zeiss, Milan, Italy) and a computer program (25). In particular, the densitometry analysis of immunocytochemistry photographs (n 5) was carried out in section in which the jejunum was oriented in order to observe all the mucosa villi. In this type of section it is possible to evaluate the presence/absence of the positive staining. Terminal deoxynucleotidyltransferase (TdT)-mediated uridine triphosphate (UTP) end labeling (TUNEL) assay was performed using a commercially available kit (Apotag, HRP kit, DBA). For diarrhea assessment all animals were checked three times daily and diarrhea was recorded. The severity of diarrhea was scored as follows: 0, no diarrhea; 1, mild diarrhea (staining of anus); 2, moderate diarrhea (staining over top of the legs and lower abdomen); 3, severe diarrhea (staining over legs and higher abdomen, often associated with continual oozing). Statistical analysis For tumor xenografts, the results were analyzed by 1-way ANOVA for multiple comparisons. Residuals were examined for normality and homogeneity of variance. Because the P value for the ANOVA F statistic was 0.001, we calculated the 95% condence intervals for the differences between the groups using the post-test Bonferroni multiple comparisons method (Primer of Biostatistics Statistical Software Program, McGraw-Hill Medical). A P value of 0.05 was considered signicant. Students t tests was used for the other statistical analyses. All statistical analyses were two-sided and results were considered to be statistically signicant at P 0.05. Figure 1. PARP-1 activity (A) and expression (B) in human colon cancer cell lines. A) Total cellular PARP-1 activity was measured in cell extracts obtained from colon cancer lines in the presence of nuclease-treated salmon testes DNA and 32 P-NAD . PARP activity was expressed as fmol 32P-NAD / g of protein and the results of ve independent experiments were presented as Box and Whisker plots. The results of statistical analysis by Students t test are as follows: HT-29 vs. Lovo, P 0.0001; HT-29 vs. HCT-15, P 0.0003; HT-29 vs. HCT-116 Chr3, P 0.19. Condence intervals (95%) were as follows: LoVo, 508 579; HT-29, 112168; HCT-15, 321 423; HCT-116, 523 647; HCT-116 Chr3, 148 218. B) Cell lysates (50 g) from the indicated colon cancer cell lines were electrophoresed and analyzed for the expression of PARP-1 or actin. The signals were quantied by densitometric scanning of the autoradiograms and normalized in relation to actin. Histograms represent the mean values of the ratios between optical densities (O.D.) of PARP-1 and those of actin obtained from 3 different experiments. Bars: sd values.

RESULTS GPI 15427 inhibits the activity of colon cancers with different levels of PARP-1 Cell lines were analyzed for PARP-1 activity (which accounts for 90% of the cellular poly(ADP-ribosyl)ation activity) and for the expression of the corresponding protein or transcript. The results indicate that activity correlated with expression of PARP-1 protein and that LoVo and HCT-116 showed the highest level of PARP-1, whereas HT-29 and HCT-116 Chr3 the lowest (Fig. 1). HT-29 and HCT-116 Chr3 lines showed also the lowest levels of PARP-1 transcript (Fig. 2). Cells were then exposed to graded concentrations of the PARP inhibitor GPI 15427 and assayed for PARP-1 activity 1 h later. The results show that GPI 15427 inhibited PARP-1 activity in all cell lines,
E4 Vol. 20 August 2006

with comparable IC50s (range of mean values: 74 87 nM, Table 2), indicating high and uniform penetration of the drug.
TENTORI ET AL.

The FASEB Journal

a 15 min treatment with GPI 15427 induced 90 2% inhibition of PARP-1 activity, which persisted unchanged 24 h later when GPI 15427 was not removed from culture medium (90 5% inhibition). On the other hand, activity largely recovered within 24 h when the inhibitor was removed from culture (23 4% inhibition). Comparable results were obtained in HT-29 cells, characterized by low levels of PARP-1 (data not shown). GPI 15427 increases the antiproliferative activity of CPT-11 and TMZ combination in colon cancer cells with different chemosensitivity proles It has been demonstrated that MR deciency is associated with resistance to TMZ and increased sensitivity to CPT-11 (9 11). All the cell lines used for this study, with the exception of HT-29 and HCT-116 Chr3, are characterized by MR functional defects (Table 1). Analysis of the expression of breast cancer resistance protein, an ATP binding cassette transporter, which is regarded as an important determinant of resistance to camptothecins (26), reveals that HT-29 expressed higher levels of this protein with respect to the other cell lines. Analysis of O6-alkylguanine DNA alkyltransferase, as an indicator of cell ability to repair O6methylguanine, and of N-methylpurine DNA glycosylase, which repairs N-methylpurines, indicates that HT-29 and LoVo cells showed the highest levels of O6-alkylguanine DNA alkyltransferase and N-methylpurine DNA glycosylase activity (Fig. 3). Evaluation of sensitivity to SN-38 reveals that HT-29 was the most resistant, in accordance with breast cancer resistance protein expression and MR prociency. The MR-procient HCT-116 Chr3 was more resistant than its MR-decient counterpart, consistent with a protective role of MR against camptothecin cytotoxicity. Most cell lines were resistant to TMZ either due to MR deciency (HCT-15, HCT-116, LoVo) or to elevated O6-alkylguanine DNA alkyltransferase activity (HT-29). The MR-procient HCT-116 Chr3 line was 3-fold more sensitive than its MR-decient counterpart. In all cell lines, 2 M GPI 15427 signicantly enhanced the antiproliferative effects of SN-38 (25-fold) or TMZ (5to 16-fold) (Fig. 4). Since TMZ and CPT-11 combination is currently evaluated in clinical trials, we then investigated the inuence on tumor cell growth of the combination TMZ SN-38 in the presence or not of the PARP inhibitor GPI 15427. Cells were treated with TMZ and SN-38 in combination at xed equipotent ratios selected on the basis of the IC50 values of each cell line, as described in Materials and Methods. Figure 5 shows the results of the median effect analysis using Calcusyn Software. The CI values indicate that the combination TMZ SN-38 was highly synergistic in LoVo cells and antagonistic in HCT-116 Chr3 cells; the addition of GPI 15427 always potentiated the antiproliferative effect of the drug combination.
E5

Figure 2. PARP-1 transcript in human colon cancer cell lines. Expression of PARP-1 was evaluated by Northern blot analysis. The hybridization signals were quantied by densitometric scanning of the autoradiograms and normalized in relation to GAPDH, actin or ethidium bromide staining of rRNA. Histograms represent the mean value of the ratios between the optical densities (O.D.) of PARP-1 and those of GAPDH, actin or ethidium bromide staining for each cell line. Bars: sd values. The results are representative of one of three different experiments.

Recovery of PARP-1 activity on GPI 15427 treatment was assessed exposing LoVo cells, which express high levels of PARP-1, to 2 M GPI 15427 (a concentration equivalent to the peak plasma concentration reached in animals treated per os with 40 mg/kg) (16). The results of three independent experiments indicate that
TABLE 2. Inhibitory effects of GPI 15427 on PARP-1 activity
95% Condence interval

Cell lines

IC50 (nM)a

HT-29 LoVo HCT-15 HCT-116 HCT-116 Chr3

87 74 86 74 79

8291 6084 7696 6585 7088

a The ability of GPI 15427 to inhibit PARP activity of colon cancer lines was assessed in cells permeabilized with digitonin and in the presence of 3H-NAD . Values represent the means of GPI 15427 IC50s calculated from 3 independent experiments. GPI 15427 IC50 on puried PARP-1 (100 ng, Alexis) was 36 nM (95% condence interval: 28 44) in accordance with previous studies (13).

PARP INHIBITOR FOR CHEMOSENSITIZATION/PROTECTION

Figure 3. Analysis of breast cancer resistance protein expression and O6-alkylguanine DNA alkyltransferase or N-methylpurine DNA glycosylase activity in human colon cancer cell lines. Cell lysates (50 g) from the indicated colon cancer cell lines were electrophoresed and analyzed for the expression of breast cancer resistance protein (BCRP) or actin (upper panel). O6-alkylguanine DNA alkyltransferase (AGT) activity is expressed as fmol of methyl groups per mg of total protein and the results of ve independent experiments were presented as Box and Whisker plots (middle panel). The results of statistical analysis by Students t test are as follows: HT-29 vs. Lovo, P 0.36; LoVo or HT-29 vs. HCT-15, HCT-116, or HCT-116 Chr3, P 0.0001. Condence intervals (95%) were as follows: LoVo, 675755; HT-29, 707775; HCT-15, 491563; HCT-116, 196 214; HCT-116 Chr3, 140 177. N-methylpurine DNA glycosylase (MPG) activity is expressed as fmol of methylpurines released per mg of protein and the results of 5 independent experiments were presented as Box and Whisker plots (lower panel). The results of statistical analysis by Students t test are as follows: HT-29 vs. LoVo, P 0.08; LoVo or HT-29 vs. HCT-15 or HCT-116 or HCT-116 Chr3, P 0.0001. Condence intervals (95%) were as follows: LoVo, 454 546; HT-29, 531 639; HCT-15, 274 356; HCT-116, 260 322; HCT-116 Chr3, 167225.
E6 Vol. 20 August 2006

Figure 4. Chemosensitivity of colon cancer cells to TMZ or SN-38 as single agents or in combination with the PARP inhibitor GPI 15427. Chemosensitivity of colon cancer cells to SN-38 or TMZ as single agents or in combination with 2 M GPI 15427 was assessed by colony formation assay and the results were expressed as IC50. Only colonies comprising 50 cells were scored as survival colonies. Data are means from three independent experiments; bars: sd. The results of statistical analysis by Students t test of the differences in sensitivity to SN-38 are as follows: HT-29 vs. LoVo, P 0.04; vs. HCT-15, P 0.03; vs. HCT-116 P 0.0001; vs. HCT-116 Chr3, P 0.04. GPI 15427 signicantly enhanced the antiproliferative effects of SN-38 or TMZ in all cell lines (P 0.0001).

Oral administration of GPI 15427 enhances the antitumor efcacy of TMZ CPT-11 in HT-29 and LoVo xenografts To examine whether GPI 15427 was capable of inhibiting PARP in vivo, mice were treated with 40 mg/kg GPI 15427 per os, a dose capable of increasing the antitumor efcacy of TMZ (16), and 1 h or 24 h later PBL were analyzed for PARP activity. The results indicate that GPI 15427 inhibited PARP activity of PBL by 60% [untreated control: fmol 23,3 (95% condence intervals: 1728), GPI 15427: fmol 9,7
TENTORI ET AL.

The FASEB Journal

Figure 5. Synergistic effect of GPI 15427 on TMZ and SN-38 combination in colon cancer cell lines. Cells were treated with TMZ SN-38 (diamond symbols) at xed molar ratios (1, 0.5, 0.2, 0.1 times the IC50 concentration for each drug) or with TMZ SN-38 and 2 M GPI 15427 (square symbols). Colony-forming ability was evaluated after 10 14 days. Combination index (CI)-fraction affected (Fa) plots of interactions between TMZ and SN-38 or TMZ, SN-38 and GPI 15427 were generated by computer analysis using CalcuSyn software Version 2.0. The combination index (CI) indicates a synergistic (CI 1), additive (CI 1) or antagonistic effect (CI 1).

(95% condence intervals: 510), P 0.0007], suggesting that the compound was bioavailable and pharmacologically active. At 24 h activity completely recovered. Toxicity studies determined the ve daily administrations of 40 mg/kg/day GPI 15427 per os 10 mg/kg/ day/i.p. TMZ 4 mg/kg/day/i.p. CPT-11 as the maximum tolerated dose of the 3 drug combination. Among the colon cancer cell lines analyzed in vitro, HT-29 and LoVo were selected for the in vivo studies. HT-29 is a highly aggressive tumor with a doubling time of 9 days and is the most resistant to SN-38 in vitro. From colon cell lines with higher PARP activities, LoVo was chosen because of its high resistance to SN-38. Moreover, both lines are resistant to TMZ due to high O6-alkylguanine DNA alkyltransferase levels (HT-29) or to MR deciency (LoVo). In HT-29 xenograft model, TMZ CPT-11 did not inhibit tumor growth. It is noteworthy that GPI 15427 signicantly enhanced the antitumor effect of TMZ CPT-11 combination (Fig. 6 and Table 3). On the other hand, treatment with GPI 15427 did not signicantly affect tumor growth inhibition in the groups treated with TMZ or CPT-11. In LoVo xenografts the combination of TMZ CPT-11 was more effective than in HT-29 xenografts, resulting in a statistically signicant delay in tumor growth. Also in this model, GPI 15427 signicantly enhanced growth inhibition induced by CPT-11 TMZ, increasing the
PARP INHIBITOR FOR CHEMOSENSITIZATION/PROTECTION

growth delay by 12 days with respect to treatment with CPT-11 TMZ and causing tumor regression in all animals (Table 3).

Figure 6. GPI 15427 per os increases the antitumor activity of CPT-11 and TMZ combination in resistant and highly proliferating HT-29 colon cancer cells. Symbols represent the means of tumor nodule volumes determined in 8 animals for each group every 3 days. Bars: sd The tumor growth curve corresponding to the group treated with GPI 15427 as a single agent was omitted since it overlaps with that of the control group. Differences between the three drug combination and all the other groups were statistically signicant starting from day 10 onward (P 0.05, using the post-test Bonferroni multiple comparisons method). Treatment produced animal body wt loss 10% of original wt. All mice recovered the initial body wt 2 wk after treatment.
E7

TABLE 3. GPI 15427 per os increases the antitumor activity of CPT-11 and TMZ combination in drug-resistant LoVo and HT-29 colon cancer cellsa
LoVob (treatment) T-C (days)c Nadir weight loss (%)d Regressionse Nadir inhibition (%)

GPI 15427 TMZ (10 mg/kg) CPT-11 (4 mg/kg) GPI 15427 TMZ GPI 15427 CPT-11 TMZ CPT-11 GPI 15427 TMZ CPT-11
HT-29b (treatment)

0 3.2 2.7 7 6.7 14.9 26

T-C (days)c

0 0 0 8 0 2 8
Nadir weight loss (%)d

0/6 0/6 0/6 1/6 0/6 1/6 6/6

Regressionse

0 14(day 6) 18(day19) 39(day13) 34(day29) 43(day43) 56(day43)

Nadir inhibition (%)

GPI 15427 TMZ (10 mg/kg) CPT-11 (4 mg/kg) GPI 15427 TMZ GPI 15427 CPT-11 TMZ CPT-11 GPI 15427 TMZ CPT-11

0 3.4 5 3.6 5 5.6 10.4

0 3 3 5 5 6 7

0/6 0/6 0/6 0/6 0/6 0/6 2/6

0 14(day 6) 16(day10) 21(day 6) 17(day10) 20(day14) 57(day24)

a Values represent the maximum percentage of tumor growth inhibition in the different treatment groups calculated with respect to vehicle-treated controls. The day after tumor challenge in which the nadir of tumor growth inhibition was observed is shown in parentheses. b In vivo doubling time of LoVo and HT-29 cells was 18 and 9 days, respectively. c Xenograft response was assessed by growth delay, calculated as TC, which represents the difference in days between the median time for tumors of treated (T) and control (C) animals to reach a volume twice that of the volume at the time of initial treatment. In both HT-29 and LoVo xenograft models, differences between the 3 drug combination and all the other groups were statistically signicant (P 0.05, using the post-test Bonferroni multiple comparisons method). In LoVo xenograft, the difference between TMZ CPT-11 and TMZ or CPT-11 was statistically signicant (P 0.05, using the post-test Bonferroni). d Body weight loss was calculated as follows: initial weight-lowest weight/initial weight 100%. e Tumor regression was dened as a reduction in the tumor volume for at least 2 consecutive measurements.

Protective effect of PARP inhibitor on CPT-11induced jejunum damage and delayed diarrhea We then investigated whether oral administration of the PARP inhibitor GPI 15427 might attenuate CPT-11 toxicity, using a well-established rat model for evaluating CPT-11-induced intestinal mucosa damage (24). Rats were treated with GPI 15427 (40 mg/kg/q2 3 days) together with 30 mg/kg/q 3 days CPT-11, a dose previously used to assess the efcacy of other compounds in preventing CPT-11 intestinal damage (24). Histological analysis of jejunum sections from rats treated with CPT-11 showed severe intestinal injury (Fig. 7A). Necro-inammatory debris in small foci was detected within the epithelial cell layer together with focal loss of cells. Numerous neutrophils were observed among epithelial cells and edematous lamina propria. Additionally, some areas with an increased inammatory reaction characterized by lymphatic inltration were detected (Fig. 7B). The severity of mucosa damage was also demonstrated by the presence of apoptotic cells (Fig. 7C) and mitotic gures (Fig. 7D). Conversely, the sections of jejunum from rats treated with GPI 15427 CPT-11 or with vehicle showed no signicant histological alterations (Fig. 7E, F). To assess whether intestinal damage was associated with PARP-1 activation, jejunum sections were analyzed for the presence of poly(ADP-ribosyl)ated proteins by immunohistochemical staining with an anti-(ADP-ribose) polymers Ab. Jejunum sections from CPT-11E8 Vol. 20 August 2006

treated rats showed positive staining for (ADP-ribose) polymers in the damaged areas (Fig. 8A) mainly localized in epithelial cells (Fig. 8B), in inammatory cells (Fig. 8C), and in the surrounding area (Fig. 8D). Densitometry analysis indicated that 6.1 0.21% of total jejunum tissue area positively stained with anti(ADP-ribose) polymers Ab (Fig. 8D), whereas only 0.5 0.09% of total jejunum tissue area was found to be positive in the jejunum of rats treated with GPI 15427 CPT-11 (Fig. 8E). No positive (ADP-ribose) polymers staining was found in the jejunum of rats treated with vehicle (Fig. 8F). To test whether the tissue damage was associated with apoptosis, we measured TUNEL-like staining in the jejunum tissue. Almost no apoptotic cells were detectable in the jejunum tissue of vehicle-treated rats (Fig. 9A), while sections from CPT-11-treated rats showed a marked appearance of dark brown apoptotic cells (Fig. 9B, B1), characterized by chromatin compaction into uniformly dense masses in perinuclear membrane, formation of apoptotic bodies and membrane blebbing (see particles B1). No apoptotic cells or fragments were found in the jejunum of rats treated with GPI 15427 CPT-11 (Fig. 9C). The ability of GPI 15427 to protect animals from delayed diarrhea was also investigated. In animals that received CPT-11 mild to moderate diarrhea developed after 24 h. Oral administration of GPI 15427 reduced severity of delayed diarrhea (Fig. 10). No diarrhea was recorded in control animals.
TENTORI ET AL.

The FASEB Journal

DISCUSSION In the present study we demonstrate for the rst time that oral administration of GPI 15427, a recently devel-

Figure 7. Protective effects of GPI 15427 on CPT-11induced jejunum damage. The histological analysis of jejunum sections stained with hematoxylin/eosin from rats treated with CPT-11 showed severe intestinal injury (A). In particular, there was necro-inammatory debris in small foci within the epithelial cell layer with focal loss of cells; numerous neutrophils were observed among epithelial cells and edematous lamina propria as well as lymphatic inltration was detected (B). The severity of mucosa damage was also demonstrated by the presence of apoptotic cells (C) and mitotic gures (D). Conversely, the sections of jejunum from rats treated with GPI 15427 CPT-11 (E) demonstrated no histological alterations or presence of apoptotic cells or/and mitotic gures. No histological alterations were found in the tissue from vehicle-treated rats (F). Figure is representative of at least 3 experiments performed on different experimental days.

GPI 15427 did not exacerbate myelotoxicity induced by full dose of TMZ We evaluated whether PARP inhibition exacerbates myelosuppression induced by TMZ, the dose-limiting side effect of this drug in patients. Mice were treated with GPI 15427 per os 40 mg/kg/day 5 days 100 mg/kg TMZ and complete blood count analyzed on day 8 and 15 post-initiation of treatment. TMZ treatment caused a 78% and 60% decline of WBC on days 8 and 15, respectively (Table 4), whereas GPI 15427 had no effect on WBC. Coadministration of TMZ and GPI 15427 did not enhance the myelosuppressive effects of TMZ. No changes in red blood cells, hemoglobin (Hb), or platelets were observed (data not shown).
PARP INHIBITOR FOR CHEMOSENSITIZATION/PROTECTION

Figure 8. Protective effects of GPI 15427 on CPT-11-induced (ADP-ribose) polymers formation. Analysis of jejunum sections obtained from CPT-11-treated rats showed positive staining for (ADP-ribose) polymers, a product of PARP-1 activation, in the damaged areas (A) mainly localized in epithelial cell (B) as well as in inammatory cells (C). In addition, some positive staining for (ADP-ribose) polymers was found in the surrounding area of the jejunum sections obtained from CPT-11-treated rats (D). In contrast, signicantly no positive (ADP-ribose) polymer staining was found in the jejunum of rats treated with GPI 15427 CPT-11 (E). No staining for (ADP-ribose) polymers in jejunum tissues was obtained in vehicle-treated rats (F). Figure is representative of at least 3 experiments performed on different experimental days.
E9

Figure 9. Protective effects of GPI 15427 on CPT-11-induced apoptosis analyzed by TUNEL assay. Almost no apoptotic cells were detectable in the jejunum tissue of vehicle-treated rats (A). On the contrary, sections from CPT-11-treated rats tissues demonstrated a marked appearance of dark brown apoptotic cells and intercellular apoptotic fragments (B, B1) associated with a specic apoptotic morphology, characterized by the compaction of chromatin into uniformly dense masses in perinuclear membrane, the formation of apoptotic bodies as well as membrane blebbing (see insert, B1). On the contrary, no apoptotic cells or fragments were found in the jejunum of rats treated with GPI 15427 CPT-11 (C). D) The positive staining in the Kit positive control tissue (normal rodent mammary gland). Figure is representative of at least 3 experiments performed on different experimental days.

anticancer drugs (29). All MR-decient colon cancer lines were resistant to TMZ, with IC50s 7- to 15-fold higher than the plasma peak concentration reached in patients ( 60 M). Resistance to TMZ of the MRprocient HT-29 line is instead due to high O6-alkylguanine DNA alkyltransferase activity. In regard to N-methylpurine DNA glycosylase levels and sensitivity to methylating agents, conicting results have been reported. In fact, embryonic stem cells derived from N-methylpurine DNA glycosylase knockout mice were more sensitive than wild-type (WT) cells, whereas Nmethylpurine DNA glycosylase-decient bone marrow cells were more resistant (30,31). Moreover, overexpression of N-methylpurine DNA glycosylase was associated with increased sensitivity to TMZ, suggesting that imbalanced BER process might generate toxic intermediates (32). Our results indicate that the ability to repair N-methylpurines does not seem to substantially affect susceptibility to TMZ. In fact, HCT-15 and HCT116, which are both MR decient and tolerant to O6-methylguanine, possess different sensitivity to TMZ even though they have similar N-methylpurine DNA glycosylase activity. MR-decient lines were more sensitive to SN-38 than the MR-procient HT-29 line and HCT-116 was more sensitive than the MR-procient counterpart HCT-116 Chr3. Hypersensitivity of MR-decient tumors to topoisomerase I inhibitor has been attributed to reduced double-strand break repair by nonhomologous endjoining (10, 11). Moreover, the endogenous expression of breast cancer resistance protein, a membrane transporter known to confer resistance to camptothecins,

oped PARP inhibitor, enhances the antitumor activity of CPT-11 and TMZ used in combination against colon cancer. It is noteworthy that GPI 15427 provides protection from intestinal damage induced by CPT-11 and did not exacerbate myelotoxicity of TMZ. The cell lines tested showed different pattern of susceptibility to CPT-11 and TMZ, mainly due to breast cancer resistance protein expression, O6-alkylguanine DNA alkyltransferase levels, or MR functional status. MR plays an important role in the maintenance of genomic stability due to its involvement in the postreplicative repair. Germline mutations of MR genes cause susceptibility to hereditary nonpolyposis colorectal cancer, which accounts for 5% of colorectal cancers, and MR defects or microsatellite instability occur in 15% of sporadic colorectal tumors (27,28). Besides being involved in genomic surveillance, MR strongly inuences tumor susceptibility or resistance to a number of
E10 Vol. 20 August 2006

Figure 10. Protective effects of GPI 15427 on irinotecaninduced delayed diarrhea. The severity of diarrhea was evaluated by observers blinded to the treatment groups and scored as follows: 0, no diarrhea; 1, mild diarrhea (staining of anus); 2, moderate diarrhea (staining over top of the legs and lower abdomen); 3, severe diarrhea (staining over legs and higher abdomen, often associated with continual oozing). No diarrhea was recorded in control animals. On the contrary, in animals that received CPT-11, no diarrhea was observed in the rst 24 h; however, mild to moderate diarrhea developed after this time point. Oral administration of GPI 15427 reduced severity of delayed diarrhea. Data are means se. of 10 rats for each group. *P 0.01 vs. CPT-11.
TENTORI ET AL.

The FASEB Journal

TABLE 4.

Effect of GPI 15427 and TMZ administration on white blood cells (WBC) in nude mice
WBC (1 103/ l) Vehicle TMZ alone GPI 15427 alone Combination

Day 8 Day 15

5.85 8.62

1.125 3.17

1.22 3.44

0.39l 0.49

6.55 7.37

0.85 1.98

2.29 2.99

1.17 1.84

contributes to the resistant phenotype of HT-29 (26, 33). It has been reported that PARP-1 might participate in colorectal carcinogenesis, since its expression was found to be increased in carcinomas with respect to the corresponding normal intestinal epithelium. Moreover, PARP-1 expression was highest in undifferentiated normal cells of intestinal crypts (34). The cell lines tested in the present study differed in PARP-1 levels, with the lowest level being expressed by the moderately differentiated HT-29 (35). In all cell lines the results of colony formation assay indicated that inhibition of PARP by GPI 15427 signicantly increased sensitivity to both SN-38 and TMZ. The entity of the chemosensitizing effect was not substantially inuenced by endogenous PARP-1 levels. In fact, GPI 15427 induced a similar sensitization to TMZ or SN-38 in LoVo and HT-29 cells, which possess different PARP-1 levels. In regard to TMZ, the most pronounced effect was observed in HCT-15 cells, which showed the highest level of resistance to the drug. In HT-29 cells, GPI 15427 enhanced TMZ antiproliferative activity by 5-fold, whereas O6benzylguanine, a selective inhibitor of O6-alkylguanine DNA alkyltransferase currently under investigation as a chemosensitizer in combination with TMZ, increased the cytostatic effect of TMZ barely by 1.2-fold (36). In addition, O6-alkylguanine DNA alkyltransferase inhibitors cannot be utilized for chemosensitization to TMZ of MR-decient tumors. Treatment with TMZ, followed by SN-38, induced synergistic antiproliferative effects mainly in LoVo cells; this effect was conrmed in vivo, since the anticancer drug combination signicantly delayed cell growth with respect to treatment with the single agents. It has been reported that O6-methylguanine induced by N-methyl-N -nitro-N-nitrosoguanidine increases topoisomerase I cleavage complexes (37). Our data indicate that the synergistic antiproliferative effect exerted by TMZ CPT-11 seems to be independent of O6-methylguanine, since all cell lines express moderate/high levels of O6-alkylguanine DNA alkyltransferase and are consistent with those previously reported in rhabdomyosarcoma and neuroblastoma models (3). In HCT116 Chr3, the TMZ and SN-38 combination resulted in antagonist effect likely due to tumor cell sensitivity to TMZ. It can be hypothesized that TMZ-induced growth arrest would prevent cells from entering S phase, thus counteracting the toxicity of topoisomerase I inhibitors, which requires DNA duplication and occurs when replication fork encounters a cleavable complex. Inhibition of PARP markedly increased the antiproPARP INHIBITOR FOR CHEMOSENSITIZATION/PROTECTION

liferative effects of the combination TMZ SN-38 in all cell lines. Moreover, the in vivo results indicate that this potentiation was particularly evident with the highly aggressive HT-29 line, which possesses a doubling time 2-fold shorter than LoVo and was less responsive to TMZ CPT-11. The results obtained in vivo also indicate that oral administration of GPI 15427 signicantly inhibited PARP activity of PBL, suggesting that the compound is well absorbed and pharmacologically active. In the experimental in vivo model used in the present study, TMZ or CPT-11 did not inhibit tumor growth nor was their activity signicantly potentiated when used in combination with the PARP inhibitor. The apparent discrepancy with the in vitro results showing the ability of GPI 15427 to enhance the antiproliferative activity of TMZ or CPT-11 combined with the PARP inhibitor is likely due to the low doses of chemotherapeutic agents used in vivo. A major concern with the use of biomodulators of resistance is the increase of toxicity of chemotherapy toward normal tissues. In regard to CPT-11, its doselimiting toxicity is diarrhea, which is of two types: acute and delayed-onset. Acute diarrhea occurs within 24 h, is the result of the cholinergic activity of CPT-11, and is prevented or rapidly suppressed by atropine. Delayedonset diarrhea occurs 24 h after drug administration and can be grade 3 (severe) or grade 4 (life-threatening) in up to 40% of the patients (38). The current strategy to treat delayed diarrhea consists in a high dose of the mu-opiate receptor agonist loperamide or somatostatin analog octreotide, but the success of these approaches is often limited (39). Delayed diarrhea has been attributed to SN-38, generated from CPT-11 by intestinal carboxylesterases or from the SN-38 glucuronide present in bile by mucosal and bacterial -glucoronidase. However, the mechanism by which this molecule provokes mucosa injury in the small intestine is still uncertain (24). In the present study we demonstrate that CPT-11 doses that generate high SN-38 concentrations induce PARP-1 overactivation with ADPribose polymer formation, which contributes to intestinal damage. Oral administration of GPI 15427 prevents ADP-ribose polymer accumulation and protects epithelial cells from cell death. In fact, an extensive synthesis of (ADP-ribose) polymers is known to deplete the cellular pools of NAD and ATP leading to cell death. PARP inhibition would spare cells from energy loss, preventing metabolic failure and providing cytoprotection (40). Indeed, GPI 15427 reduced inammation of jejunum and severity of delayed diarrhea in animals
E11

receiving CPT-11 at doses higher than those used in combination with TMZ. The involvement of PARP-1 in inammatory diseases, including acute and chronic inammation of the gut, and the protective effect of PARP inhibitors have been demonstrated in various experimental models (41). Besides the involvement in PARP-mediated cellular suicidal pathway, PARP-1 enhances the activities of NF- B and activating protein AP-1, key transcription factors regulating the expression of inammatory mediators and adhesion molecules (41). In the present model, the PARP inhibitor also prevents inammation as indicated by the preservation of the intestinal architecture, with the absence of edema and lymphatic inltration. Reduction of CPT-11 toxicity by PARP inhibitor appears in some ways in contrast to the enhancement of antitumor efcacy exerted by GPI 15427 in combination with TMZ CPT-11. This dual role of PARP inhibitor is due to the fact that PARP-1 may act as a survival factor, favoring DNA repair when the damage is moderate, or as a cell death mediator in the presence of extensive DNA damage that triggers PARP-1 overactivation. In this case, oral doses of GPI 15427 allow immediate achievement of sufcient PARP inhibition in the gut wall just where it is required to prevent mucosal damage. On the other hand, chemosensitization induced at the tumor level by GPI 15427 in combination with low doses of CPT-11 TMZ relies on inhibition of the PARP activity that is needed for DNA repair. GPI 15427 did not exacerbate myelotoxicity induced by full doses of TMZ. This might be due to the complete recovery of PARP function when GPI 15427 is no longer available. On the contrary, when TMZ is combined in vivo with O6-benzylguanine, an increase of myelotoxicity was reported, which required a marked reduction of TMZ dosing (42). Indeed, when HT-29 or LoVo cells were treated with O6-benzylguanine O6alkylguanine DNA alkyltransferase activity did not recover after 24 h (data not shown). Upon interaction with O6-benzylguanine, O6-alkylguanine DNA alkyltransferase is rapidly degraded by the proteasome, and this contributes to induce a long-lasting depletion of O6-alkylguanine DNA alkyltransferase since the protein has to be resynthesized (43). The ability of GPI 15427 to prevent CPT-11 intestinal toxicity but not TMZinduced myelotoxicity might be attributed to the higher concentrations of the PARP inhibitor reached in the gut by means of oral administration with respect to those reached in bone marrow following absorption and distribution of the compound. In conclusion, these data indicate that GPI 15427 sensitizes tumors to methylating agents and topoisomerase I poison combination, representing a novel strategy to enhance the efcacy and reduce toxicity of chemotherapy in colon cancer. Since PARP inhibitors have recently entered phase I-II clinical trials in combination with TMZ, our ndings on the combination of PARP inhibitor with TMZ CPT-11 will provide the rational basis for the development of new clinical protocols.
E12 Vol. 20 August 2006

This work was supported by grants from the Italian Ministry of Education and Research Fondo per gli Investimenti della Ricerca di Base (FIRB) to G.G. and Programmi di Ricerca scientica di rilevante Interesse Nazionale (PRIN) projects to G.G. and L.T. J.Z., W.X., and R.L. work for MGI PHARMA, which is developing PARP inhibitors for cancer treatment.

REFERENCES
1. Douillard, J. Y., Cunningham, D., Roth, A. D., Navarro, M., James, R. D., Karasek, P., Jandik, P., Iveson, T., Carmichael, J., Alakl, M., et al. (2000) Irinotecan combined with uorouracil compared with uorouracil alone as rst-line treatment for metastatic colorectal cancer: a multicentre randomised trial. Lancet 355, 10411047 Rasheed, Z. A., and Rubin, E. H. (2003) Mechanisms of resistance to topoisomerase I-targeting drugs. Oncogene 22, 7296 7304 Houghton, P. J., Stewart, C. F., Cheshire, P. J., Richmond, L. B., Kirstein, M. N., Poquette, C. A., Tan, M., Friedman, H. S., and Brent, T. P. (2000) Antitumor activity of temozolomide combined with irinotecan is partly independent of O6-methylguanine-DNA methyltransferase and mismatch repair phenotypes in xenograft models. Clin. Cancer Res. 6, 4110 4118 Patel, V. J., Elion, G. B., Houghton, P. J., Keir, S., Pegg, A. E., Johnson, S. P., Dolan, M. E., Bigner, D. D., and Friedman, H. S. (2000) Schedule-dependent activity of temozolomide plus CPT-11 against a human central nervous system tumor-derived xenograft. Clin. Cancer Res. 6, 4154 4157 Jones, S. F., Gian, V. G., Greco, F. A., Miranda, F. T., Shipley, D. L., Thompson, D. S., Hainsworth, J. D., Toomey, M. A., Willcutt, N. T., and Burris, H. A., 3rd. (2003) Trial of irinotecan and temozolomide in patients with solid tumors. Oncology 17, 41 45 Wagner, L. M., Crews, K. R., Iacono, L. C., Houghton, P. J., Fuller, C. E., McCarville, M. B., Goldsby, R. E., Albritton, K., Stewart, C. F., and Santana, V. M. (2004) Phase I trial of temozolomide and protracted irinotecan in pediatric patients with refractory solid tumors. Clin. Cancer Res. 10, 840 848 Reardon, D. A., Quinn, J. A., Rich, N. J., Desjardins, A., Vredenburgh, J., Gururangan, S., Sathornsumetee, S., Badruddoja, M., McLendon, R., Provenzale, J., et al. (2005) Phase I trial of irinotecan plus temozolomide in adults with recurrent malignant glioma. Cancer 104, 1478 1486 Tentori, L., and Graziani, G. (2004) Temozolomide: an update on pharmacological strategies to increase its antitumor activity. Med. Chem. Rev.-Online. 1, 144 150 DAtri, S., Tentori, L., Lacal, P. M., Graziani, G., Pagani, E., Benincasa, E., Zambruno, G., Bonmassar, E., and Jiricny, J. (1998) Involvement of the mismatch repair system in temozolomide-induced apoptosis. Mol. Pharmacol. 54, 334 341 Jacob, S., Aguado, M., Fallik, D., and Praz, F. (2001) The role of the DNA mismatch repair system in the cytotoxicity of the topoisomerase inhibitors camptothecin and etoposide to human colorectal cancer cells. Cancer Res. 61, 6555 6562 Magrini, R., Bhonde, M. R., Hanski, M. L., Notter, M., Scherubl, H., Boland, C. R., Zeitz, M., and Hanski, C. (2002) Cellular effects of CPT-11 on colon carcinoma cells: dependence on p53 and hMLH1 status. Int. J. Cancer 101, 2331 Tentori, L., Leonetti, C., Scarsella, M., dAmati, G., Portarena, I., Zupi, G., Bonmassar, E., and Graziani, G. (2002) Combined treatment with temozolomide and poly(ADP-ribose) polymerase inhibitor enhances survival of mice bearing hematologic malignancy at the central nervous system site. Blood 99, 2241 2244 Tentori, L., Leonetti, C., Scarsella, M., DAmati, G., Vergati, M., Portarena, I., Xu, W., Kalish, V., Zupi, G., Zhang, J., and Graziani, G. (2003) Systemic administration of GPI 15427, a novel poly(ADP-ribose) polymerase-1 inhibitor, increases the antitumor activity of temozolomide against intracranial melanoma, glioma, lymphoma. Clin. Cancer Res. 9, 5370 5379 Miknyoczki, S. J., Jones-Bolin, S., Pritchard, S., Hunter, K., Zhao, H., Wan, W., Ator, M., Bihovsky, R., Hudkins, R., Chatterjee, S., et al. (2003) Chemopotentiation of temozolomide, irinotecan,

2. 3.

4.

5.

6.

7.

8. 9.

10.

11.

12.

13.

14.

The FASEB Journal

TENTORI ET AL.

15.

16.

17.

18.

19. 20. 21. 22.

23.

24.

25.

26.

27. 28.

29. 30.

31. 32.

and cisplatin activity by CEP-6800, a poly(ADP-ribose) polymerase inhibitor. Mol. Cancer Ther. 2, 371382 Calabrese, C. R., Almassy, R., Barton, S., Batey, M. A., Calvert, A. H., Canan-Koch, S., Durkacz, B. W., Hostomsky, Z., Kumpf, R. A., Kyle, S., et al. (2004) Anticancer chemosensitization and radiosensitization by the novel poly(ADP-ribose) polymerase-1 inhibitor AG 14361. J. Natl. Cancer Inst. 96, 56 67 Tentori, L., Leonetti, C., Scarsella, M., Vergati, M., Xu, W., Calvin, D., Morgan, L., Tang, Z., Woznizk, K., Alemu, C., et al. (2005) Brain distribution and efcacy as chemosensitizer of an oral formulation of PARP-1 inhibitor GPI 15427 in experimental models of CNS tumors. Int. J. Oncol. 26, 415 422 Schreiber, V., Ame, J. C., Dolle, P., Schultz, I., Rinaldi, B., Fraulob, V., Menissier-de Murcia, J., and de Murcia, G. (2002) Poly(ADP-ribose) polymerase-2 (PARP-2) is required for efcient base excision DNA repair in association with PARP-1 and XRCC1. J. Biol. Chem. 277, 23028 23036 Yung, T. M., Sato, S., and Satoh, M. S. (2004) Poly(ADPribosyl)ation as a DNA damage-induced post-translational modication regulating poly(ADP-ribose) polymerase-1 topoisomerase I interaction. J. Biol. Chem. 279, 39686 39696 Malanga, M., and Althaus, F. R. (2004) Poly(ADP-ribose) reactivates stalled DNA topoisomerase I and induces DNA strand break resealing. J. Biol. Chem. 279, 5244 5248 Tentori, L., and Graziani, G. (2005) Chemopotentiation by PARP inhibitors in cancer therapy. Pharmacol. Res. 52, 2533 Ha, H. C., and Snyder, S. H. (1999) Poly(ADP-ribose) polymerase is a mediator of necrotic cell death by ATP depletion. Proc. Natl. Acad. Sci. U. S. A. 96, 13978 13982 Tentori, L., Vergati, M., Muzi, A., Levati, L., Rufni, F., Forini, O., Vernole, P., Lacal, P. M., and Graziani, G. (2005) Generation of an immortalized human endothelial cell line as a model of neovascular proliferating endothelial cells to assess chemosensitivity to anticancer drugs. Int. J. Oncol. 27, 525535 Bakondi, E., Bai, P., Szabo, E. E., Hunyadi, J., Gergely, P., Szabo, C., and Virag, L. (2002) Detection of poly(ADP-ribose) polymerase activation in oxidatively stressed cells and tissues using biotinylated NAD substrate. J. Histochem. Cytochem. 50, 9198 Fittkau, M., Voigt, W., Holzhausen., H. J., and Schmoll, H. J. (2004) Saccharic acid 1.4-lactone protects against CPT-11-induced mucosa damage in rats. J. Cancer Res. Clin. Oncol. 130, 388 394 Cuzzocrea, S., Misko, T. P., Costantino, G., Mazzon, E., Micali, A., Caputi A. P., Macarthur, H., and Salvemini, D. (2000) Benecial effects of peroxynitrite decomposition catalyst in a rat model of splanchnic artery occlusion and reperfusion. FASEB J. 14, 10611072 Rajendra, R., Gounder, M. K., Saleem, A., Schellens, J. H., Ross, D. D., Bates, S. E., Sinko, P., and Rubin, E. H. (2003) Differential effects of the breast cancer resistance protein on the cellular accumulation and cytotoxicity of 9-aminocamptothecin and 9-nitrocamptothecin. Cancer Res. 63, 3228 3233 Rowley, P. T. Inherited susceptibility to colorectal cancer. (2005) Annu. Rev. Med. 56, 539 554 Herman, J. G., Umar, A., Polyak, K., Graff, J. R., Ahuja, N., Issa, J. P., Markowitz, S., Willson, J. K., Hamilton, S. R., Kinzler, K. W., et al. (1998) Incidence and functional consequences of hMLH1 promoter hypermethylation in colorectal carcinoma. Proc. Natl. Acad. Sci. U. S. A. 95, 6870 6875 Fedier, A., and Fink, D. (2004) Mutations in DNA mismatch repair genes: implications for DNA damage signaling and drug sensitivity. Int. J. Oncol. 24, 1039 1047 Engelward, B. P., Weeda, G., Wyatt, M. D., Broekhof, J. L., de Wit, J., Donker, I., Allan, J. M., Gold, B., Hoeijmakers, J. H., and Samson, L. D. (1997) Base excision repair decient mice lacking the Aag alkyladenine DNA glycosylase. Proc. Natl. Acad. Sci. U. S. A. 94, 1308713092 Roth, R. B., and Samson, L. D. (2002) 3-Methyladenine DNA glycosylase-decient Aag null mice display unexpected bone marrow alkylation resistance. Cancer Res. 62, 656 660 Trivedi, R. N., Almeida, K. H., Fornsaglio, J. L., Schamus, S., and Sobol, R. W. (2005) The role of base excision repair in the

33. 34.

35. 36.

37.

38.

39.

40.

41. 42.

43. 44.

45.

46.

47.

48.

sensitivity and resistance to temozolomide-mediated cell death. Cancer Res. 65, 6394 6400 Doyle, L. A., and Ross, D. D. (2003) Multidrug resistance mediated by the breast cancer resistance protein breast cancer resistance protein (ABCG2). Oncogene 22, 7340 7358 Idogawa, M., Yamada, T., Honda, K., Sato, S., Imai, K., and Hirohashi, S. (2005) Poly(ADP-ribose) polymerase-1 is a component of the oncogenic T-cell factor-4/beta-catenin complex. Gastroenterology 128, 1919 1936 Fogh, J., and Trempe, G. (1975) New human tumour cell lines. In Human tumour cells in vitro (Fogh, J., ed) pp. 115141, Plenum Press, New York, U. S. A. Wedge, S. R., Porteus, J. K., May, B. L., and Newlands, E. S. (1996) Potentiation of temozolomide and BCNU cytotoxicity by O(6)-benzylguanine: a comparative study in vitro. Br. J. Cancer 73, 482 490 Pourquier, P., Waltman, J. L., Urasaki, Y., Loktionova, N. A., Pegg, A. E., Nitiss, J. L., and Pommier, Y. (2001) Topoisomerase I-mediated cytotoxicity of N-methyl-N -nitro-N-nitrosoguanidine: trapping of topoisomerase I by the O6-methylguanine. Cancer Res. 61, 5358 Saliba, F., Hagipantelli, R., Misset, J. L., Bastian, G., Vassal, G., Bonnay, M., Herait, P., Cote, C., Mahjoubi, M., Mignard, D., and Cvitkovic, E. (1998) Pathophysiology and therapy of irinotecaninduced delayed-onset diarrhea in patients with advanced colorectal cancer: a prospective assessment. J. Clin. Oncol. 16, 27452751 Benson, A. B., 3rd, Ajani, J. A., Catalano, R. B., Engelking, C., Kornblau, S. M., Martenson, J. A., Jr., McCallum, R., Mitchell, E. P., ODorisio, T. M., Vokes, E. E., and Wadler, S. (2004) Recommended guidelines for the treatment of cancer treatment-induced diarrhea. J. Clin. Oncol. 22, 2918 2926 Mazzon, E., Dugo, L., Li, J. H., Di Paola, R., Genovese, T., Caputi, A. P., Zhang, J., and Cuzzocrea, S. (2002) GPI 6150, a PARP inhibitor, reduces the colon injury caused by dinitrobenzene sulfonic acid in the rat. Biochem. Pharmacol. 64, 327337 Cuzzocrea, S. (2005) Shock, inammation and PARP. Pharmacol. Res. 52, 72 82 Quinn, J. A., Desjardins, A., Weingart, J., Brem, H., Dolan, M. E., Delaney, S. M., Vredenburgh, J., Rich, J., Friedman, A. H., Reardon, D. A., et al. (2005) Phase I trial of temozolomide plus O6-benzylguanine for patients with recurrent or progressive malignant glioma. J. Clin. Oncol. 23, 7178 7187 Xu-Welliver, M., and Pegg, A. E. (2002) Degradation of the alkylated form of the DNA repair protein, O(6)-alkylguanineDNA alkyltransferase. Carcinogenesis 23, 823 830 Umar, A., Boyer, J. C., Thomas, D. C., Nguyen, D. C., Risinger, J. I., Boyd, J., Ionov, Y., Perucho, M., and Kunkel, T. A. (1994) Defective mismatch repair in extracts of colorectal and endometrial cancer cell lines exhibiting microsatellite instability. J. Biol. Chem. 269, 1436714370 Gibson, N. W., Hartley, J. A., Barnes, D., and Erickson, L. C. (1986) Combined effects of streptozotocin and mitozolomide against four human cell lines of the Mer phenotype. Cancer Res. 46, 4995 4998 Papadopoulos, N., Nicolaides, N. C., Liu, B., Parsons, R., Lengauer, C., Palombo, F., DArrigo, A., Markowitz, S., Willson, J. K., Kinzler, K. W., et al. (1995) Mutations of GTBP in genetically unstable cells. Science 268, 19151917 Papadopoulos, N., Nicoladesi, N. C., Wei, Y. F., Ruben, S. M., Carter, K. C., Rosen, C. A., Haseltine, W. A., Fleischmann, R. D., Fraser, C. M., Adams, M. D., et al. (1994) Mutation of a MutL homolog in hereditary colon cancer. Science 263, 16251629 Koi, M., Umar, A., Chaulan, D. P., Cherian, S. P., Carethers, J. M., Kunkel, T. A., and Boland, C. R. (1994) Human chromosome 3 corrects mismatch deciency and microsatellite instability and reduces N-methyl-N -nitrosoguanidine tolerance in colon tumor cells with homozygous hMLH1 mutation. Cancer Res. 54, 4308 4312 Received for publication February 16, 2006. Accepted for publication March 31, 2006.

PARP INHIBITOR FOR CHEMOSENSITIZATION/PROTECTION

E13