Efficient Deletion of Normal Brca2-Deficient Intestinal Epithelium

by Poly(ADP-Ribose) Polymerase Inhibition Models Potential

Prophylactic Therapy

Trevor Hay,1Helen Jenkins,

2Owen J. Sansom,

1Niall M.B. Martin,

2

Graeme C.M. Smith,2and Alan R. Clarke

1

1School of Bioscience, Cardiff University, Cardiff, United Kingdom and 2KuDOS Pharmaceuticals Ltd., Cambridge, United Kingdom

Abstract

The genes encoding the BRCA1 and BRCA2 tumor suppressorsare the most commonly mutated in human familial breastcancers. Both have separate roles in the maintenance ofgenomic stability through involvement in homologous recom-bination, an error-free process enabling cells to repair DNAdouble-strand breaks. We have previously shown that cre-mediated conditional deletion of Brca2 within the mousesmall intestine sensitizes the tissue to DNA damage. Eventu-ally, the tissue repopulates via stem cells in which recombi-nation at the floxed Brca2 allele has not taken place. In thisstudy, we have treated Brca2-deficient small intestine with apotent small-molecule inhibitor of poly(ADP-ribose) polymer-ase 1 (PARP1), an enzyme predominantly involved in therecognition of DNA single-strand breaks. Brca2 deficiencyrendered otherwise normal cells exquisitely sensitive to PARPinhibition, resulting in very high levels of apoptosis as early as6 hours after treatment, with evidence for repopulation of thetissue at 12 hours. Furthermore, the intestines of animalstreated with serial injections of the inhibitor repopulated veryrapidly in comparison with those from untreated mice. Ourresults represent the first in vivo demonstration thatinhibition of PARP1 activity confers exquisite sensitivity todeath in physiologically normal Brca2-deficient cells, suggest-ing that such a regimen may be extremely potent prophylac-tically in women heterozygous for the BRCA2 gene, as well asagainst established tumors lacking functional BRCA2. (CancerRes 2005; 65(22): 10145-8)

Introduction

It is currently estimated that familial cases account for f10% ofhuman breast cancers (1). Of these, up to half are believed to bedue to deficiency in either of the tumor suppressors BRCA1 orBRCA2 (1). The identification of new drug regimens against suchtumors is therefore of paramount importance in the current fightagainst the disease. BRCA1 and BRCA2 have separate roles in thehomologous recombination pathway, a process by which DNAdouble-strand breaks are conservatively repaired (2). Cells lackingeither protein are prone to genomic instability due to thealternative employment of the error-prone nonhomologous end-joining pathway of DNA repair (2). Cells lacking Brca2 have been

shown to be sensitive to a variety of DNA damaging agents, eitherin vitro or in vivo , particularly following treatment with DNA cross-linking agents such as mitomycin C (3–11).The enzyme poly(ADP-ribose) polymerase 1 (PARP1) is activated

by binding to DNA strand breaks (12) and is known to have a varietyof roles, including participation in base excision repair (13). Throughits ability to modify proteins by the synthesis and elongation of ADP-ribose polymers, PARP1 reduces cellular pools of ATP and NAD+,thereby affecting many energy-dependent processes (13). As PARP1is known to actively participate in DNA repair, inhibition of itsactivity can lead to enhanced cell death either alone or in combi-nation with DNA damaging agents (12). As such, PARP1 inhibitorshave been discussed as potential chemotherapeutic agents, eitheralone or by potentiating other cytotoxic treatments (12–15). Indeed,studies have already shown the effectiveness of PARP inhibition incombination with either radiotherapy or chemotherapy in a range ofhuman tumor mouse xenograft models (16, 17). We have previouslyshown that conditional deletion of the Brca2 gene, using a well-characterized CYP1A1-driven cre-loxP approach (18, 19), sensitizesthe mouse small intestine to DNA damage and eventually leads torepopulation by stem cells in which recombination of the floxedBrca2 allele has not taken place (11). In this study, we have used ourmodel Brca2-deficient system to analyze whether PARP inhibitionfurther sensitizes the intestine to DNA damage in vivo . We show thatapoptosis is dramatically increased and that stem cell repopulationproceeds very rapidly compared with untreated tissue, showing thatcells lacking Brca2 are indeed highly sensitive to PARP inhibition. Assuch, we concur with previous studies (12–17) which suggest thattreatment with PARP inhibitors may be extremely potent againsthuman tumors in which BRCA2 is mutated. Based on our currentdata, we now propose that PARP inhibition may also be consideredas a potential prophylactic treatment of nonneoplastic BRCA2-deficient cells in women heterozygous for BRCA2, as well as aprimary follow-up treatment following surgery in patients in whomBRCA2 deficiency has led to tumorigenesis.

Materials and Methods

Experimental mice. Mice carrying the floxed Brca2 allele (Brca2 fl) were

provided by Tak Mak at the University of Toronto (Toronto, Ontario, Canada)and details of these mice can be found in ref. 11. Mice bearing the intestinal

inducible Cre vector, Ah-cre , were provided by Douglas Winton at Cambridge

University (Cambridge, United Kingdom; ref. 18). Mice carrying the Rosa26R

transgene were already in house and details can be found in ref. 20. Mice werefed standard diet and water ad libitum and all animal experiments were

carried out in accordance with current UK Home Office regulations.

Preparation and injection of DNA damaging drugs. h-Napthoflavonewas prepared as previously described (18). Mice were given an i.p. injection

of 80 mg/kg four times over 4 successive days. KU0058948 was provided by

KuDOS Pharmaceuticals (Cambridge, United Kingdom; see Fig. 1A for

Note: O. J. Sansom is currently at Cancer Research UK Beatson Laboratories,Garscube Estate, Switchback Road, Glasgow G61 1BD, United Kingdom.

Requests for reprints: Alan R. Clarke, School of Bioscience, Cardiff University,Museum Avenue, Cardiff CF10 3US, United Kingdom. Phone: 44-02920-874609; Fax: 44-02920-874116; E-mail: clarkear@Cardiff.ac.uk.

I2005 American Association for Cancer Research.doi:10.1158/0008-5472.CAN-05-1186

www.aacrjournals.org 10145 Cancer Res 2005; 65: (22). November 15, 2005

Priority Report

Research. on June 27, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

Research. on June 27, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

Research. on June 27, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

structure). It is a novel and very potent small-molecule inhibitor of both

PARP1 and PARP2, exhibiting an IC50 of 3.4 nmol/L against PARP1 and an

IC50 of 6 nmol/L against cellular PARP activity (14). KU0058948 wasformulated at 2 mg/mL in normal saline (0.9% NaCl) with heating to 80jCfor 15 minutes and subsequently stored at �20jC for up to a week. Mice

were weighed immediately before injection and administered with an i.p.

dose of 15 mg/kg.Bioavailability of poly(ADP-ribose) polymerase inhibitor in small

intestine. After a single i.p. injection of 15 mg/kg KU0058948, mice were

killed and sections of small intestine were removed. Samples were

homogenized for 30 seconds in 3 volumes of ice-cold extraction buffer[1 tablet of Complete EDTA-free protease inhibitor cocktail (Roche) in

20 mL 1% NP40/1� PBS], incubated on ice for 5 minutes, and snap frozen

on dry ice. Following thawing, the drug was extracted from the homogenateby protein precipitation with acetonitrile and supernatant was isolated

by centrifugation at 13,000 rpm for 10 minutes. Samples were analyzed by

liquid chromatography-tandem mass spectrometry (LC-MS/MS) with the

chromatography done using a Phenomenex Synergi MaxRP (50 � 2 mm)column with a gradient elution from 5% acetonitrile:95% formic acid

(0.01%) to 95% acetonitrile:5% formic acid (0.01%) over a 6.5-minute period.

The mass spectrometer used was a Sciex 2000 with TurboIonSpray

ionization, operating in positive ion mode and using multiple reactionmonitoring of ions 381.1!281.3. Calibration standards (20 ng/mL-20 Ag/mL)

were prepared in mouse plasma as a proxy matrix.

Removal and preparation of small intestines. All small intestinalpreparations, including whole mounts, were carried out as previously

described (18).

PCR for the recombined Brca2 allele. Recombined Brca2 PCR was

done on DNA from small intestinal tissue as previously described (11).Statistical analysis. All statistical analyses were done using nonpara-

metric Mann-Whitney test with a confidence level of 95%.

Results and Discussion

Bioavailability of poly(ADP-ribose) polymerase inhibitor insmall intestine after a single i.p. injection of 15 mg/kg. Toascertain the levels of KU0058948 in the small intestine after asingle i.p. dose of 15 mg/kg, LC-MS/MS analysis for the presence ofthe PARP inhibitor was done on tissue extracts from duplicate micekilled over a 24-hour period following treatment. As shown inFig. 1B , high levels of the molecule were detected 1 hour aftertreatment, after which the inhibitor was cleared relatively quicklyfrom the tissue. It should be noted that a concentration of 100 ng/mL equates to f250 nmol/L, which has been shown to beextremely effective at killing Brca2�/� mouse embryonic stem cellsin vitro (14). It was therefore clear that the PARP inhibitor waspresent in the small intestine at high levels almost immediatelyfollowing dosing and was retained at an effective level for at least12 hours. On the basis of this data, we assessed the apoptoticresponse at both 6 and 12 hours after treatment.Increased apoptosis following treatment with 15 mg/kg

poly(ADP-ribose) polymerase inhibitor. Mice were killed 6 or12 hours after a single i.p. dose of 15 mg/kg KU0058948 and theirintestines scored for apoptotic bodies and mitotic figures. At 6 hours,the levels of apoptosis had increased significantly in mice treatedwith the PARP inhibitor compared with control mice that had beeninjected with saline or were untreated (Fig. 2A ; P = 0.04). In addition,the levels of mitosis in the experimental mice were significantlyreduced (Fig. 2A ; P = 0.04), presumably as a reflection of the reducednumbers of viable cells available to drive proliferation in the affectedcrypts. At 12 hours, the level of apoptosis had increased to around35% in Brca2-deficient mice, which is highly significant whencompared with controls (Fig. 2B ; P = 0.0004). This is a high level ofapoptosis, and the representative image in Fig. 2C shows that themajority of apoptotic bodies were confined to the stem cellcompartment and the proliferative zone in the crypts of these mice.This level of apoptosis seems to reflect the peak response as levelswere significantly reduced to f13% by 24 hours (P = 0.014; data notshown). At 12 hours, we also tested the levels of apoptosis in miceheterozygous for Brca2 and found that although there was asignificant increase in apoptotic levels at this time point comparedwith untreated controls (Fig. 2B ; P = 0.015), the level of increase wasvery small compared with the induction of death in thehomozygotes. These data show that the PARP inhibitor–induceddeath was highly specific to the Brca2�/� cells.Close inspection of intestinal sections from Brca2-deficient mice

revealed subtle evidence that the tissue was starting to repopulatewith unrecombined cells at 12 hours after inhibitor treatment.Figure 2D shows a representative image from a Brca2-deficientintestine treated with the inhibitor and shows a crypt which containsvery few apoptotic bodies but many mitotic figures (arrow). Theseapparently healthy crypts contrasted the many unhealthy or ‘‘dead’’crypts, which contained many apoptotic bodies and virtually nomitotic figures (Fig. 2C and D). These healthy crypts reflect twopossibilities: First, that there would be some crypts in which cre-mediated recombination at the floxed Brca2 allele had not takenplace such that they retained both functional Brca2 alleles and sowere refractory to the effects of the PARP inhibitor. Second, thatdeletion of Brca2 in combination with PARP inhibition leads to therapid repopulation of the tissue by unrecombined cells, a processreflected in the high levels of proliferation observed.Rapid repopulation of the small intestine following serial

treatment with 15 mg/kg poly(ADP-ribose) polymerase inhib-itor. We have previously shown that despite the increased levels of

Figure 1. A, molecular structure of PARP inhibitor KU0058948. B, time courseshowing the levels of KU0058948 in mouse small intestine following a singlei.p. dose of 15 mg/kg.

Cancer Research

Cancer Res 2005; 65: (22). November 15, 2005 10146 www.aacrjournals.org

Research. on June 27, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

apoptosis after deletion of Brca2 , the small intestine repopulatesvery slowly, with f50% of recombined stem cells replaced by thosewhich had failed to recombine 6 months after Brca2 deletion (11).To test whether treatment with the PARP inhibitor increased therate of repopulation in Brca2-deficient tissue, we used mice whichcarried the Rosa26R reporter allele (20) as well as the Ah-Cre andfloxed Brca2 genes. Mice were subjected to twice-weekly injectionsof 15 mg/kg KU0058948 for 3 weeks and then scored forrecombined crypts using a simple whole-mount stain for thereporter allele (11). Figure 3A shows representative areas of smallintestine from mice, either heterozygous or homozygous for Brca2 ,treated with the PARP inhibitor or saline, and clearly shows areduction in recombined stem cells in Brca2-deficient intestine inwhich PARP inhibition has taken place. The intestines of three mice

from each treatment were scored for the percentage of recombinedcrypts (blue) and the resulting graph is shown in Fig. 3B . Signi-ficantly fewer recombined crypts remain in Brca2-deficient micetreated with the PARP inhibitor compared with untreated controls(P = 0.04). As we have previously shown (11), a simple PCR strategyconfirmed that the recombined (‘‘white’’) areas of intestinecontained no recombined floxed Brca2 allele (data not shown).Overall, the presented data report two phenomena: First, they

confirm our previous data indicating that Brca2-deficient cells,including stem cells, are removed from the small intestinefollowing DNA damage, and that this effectively prevents theaccumulation of cells with the potential to acquire further cancer-causing mutations (11). In this study, where the detection andsubsequent repair of DNA damage is reduced through inhibition of

Figure 2. Apoptotic response of Brca2 -deficient small intestine following a single i.p. injection of 15 mg/kg KU0058948. A and B, levels of apoptosis and mitosis 6 (A )or 12 (B ) hours after treatment. Open columns, apoptosis; closed columns, mitosis; y axis, percentage of apoptotic bodies/mitotic figures per crypt. C, representativeimages of small intestinal crypts from Ah-cre+ Brca2 fl/fl mice 12 hours after treatment with either saline or 15 mg/kg PARP inhibitor. Bar, 50 Am. D, representativeimage from Ah-cre+ Brca2 fl/fl small intestine 12 hours after treatment with 15 mg/kg PARP inhibitor. Arrow, a healthy, highly proliferating crypt. Bar, 50 Am.

Brca2 Deficiency Sensitizes Cells to PARP Inhibition

www.aacrjournals.org 10147 Cancer Res 2005; 65: (22). November 15, 2005

Research. on June 27, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

one of the key proteins involved in damage recognition (12), boththe apoptotic response and the rate of repopulation are increased,presumably due to the increased number of cells in which thedefective homologous recombination pathway is required butunavailable. Second, we have modeled for the first time the in vivoconsequences of drug challenge on otherwise physiologicallynormal Brca2-deficient cells. We observe highly efficient andselective deletion of Brca2�/� cells with no apparent deleteriouseffect on surrounding Brca2-functional cells or whole animalphysiology. It is interesting to note that the effects of the PARPinhibitor seem to be more highly specific to cells lacking Brca2compared with other DNA-damaging agents we have tested inour model system. Thus, we have previously shown that a lowconcentration of mitomycin C (0.1 mg/kg) specifically killsBrca2 -deficient cells in the small intestine, but at higherconcentrations this specificity was reduced, with levels of celldeath also increased in cells with functional Brca2 (11). In contrast,our current data show that PARP inhibition delivers even higherlevels of Brca2�/� cell deletion and yet has no deleterious effecton Brca2+/� (wild-type) cells.Taken together, our results go beyond previous studies (12–17)

to show that PARP inhibition is highly effective at specificallydeleting nonneoplastic Brca2-deficient cells, and thus may beeffective for prophylactic therapy or therapy subsequent to surgeryin BRCA2 mutation carriers, as well as against BRCA2-deficienttumors as previously suggested. Deficiency of Brca2 in our modelintestinal system does not predispose to tumorigenesis (11) andthus we cannot address whether Brca2-deficient tumors regress,or can be prevented, in this system. We are now pursuing studieswithin the mouse mammary epithelium which will allow usto directly test the efficacy of PARP inhibitors for both tumorand prophylactic therapy.

Acknowledgments

Received 4/6/2005; revised 9/1/2005; accepted 9/21/2005.Grant support: American Institute for Cancer Research grant no. 03-336, KuDOS

Pharmaceuticals Ltd., and Wales Gene Park.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 accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.

Figure 3. Rapid repopulation in Brca2 -deficient small intestine followingserial injections of 15 mg/kg KU0058948. A, representative fields from smallintestines of mice following six biweekly injections of PARP inhibitor. Theintestines were inverted to aid counting of positive crypts (blue ). Bar, 500 Am.B, percentage of recombined crypts (blue ) in mice treated with six biweeklyinjections of PARP inhibitor.

Cancer Research

Cancer Res 2005; 65: (22). November 15, 2005 10148 www.aacrjournals.org

References1. Pellegrini L, Venkitaraman A. Emerging functions ofBRCA2 in DNA recombination. Trends Biochem Sci2004;29:310–6.

2. Powell SN, Kachnic LA. Roles of BRCA1 and BRCA2 inhomologous recombination, DNA replication fidelityand the cellular response to ionizing radiation. Onco-gene 2003;22:5784–91.

3. Sharan SK, Morimatsu M, Albrecht U, et al. Embryoniclethality and radiation hypersensitivity mediated byRad51 in mice lacking Brca2. Nature 1997;386:804–10.

4. Donoho G, Brenneman MA, Cui TX, et al. Deletion ofBrca2 exon 27 causes hypersensitivity to DNA cross-links, chromosomal instability, and reduced life span inmice. Genes Chromosomes Cancer 2003;36:317–31.

5. D’Andrea AD, Grompe M. The Fanconi anaemia/BRCApathway. Nat Rev Cancer 2003;3:23–34.

6. Chen P-L, Chen C-F, Chen Y, Xiao J, Sharp ZD, LeeW-H. The BRC repeats in BRCA2 are critical for RAD51binding and resistance to methyl methanesulfonatetreatment. Proc Natl Acad Sci U S A 1998;95:5287–92.

7. Abbott DW, Freeman ML, Holt JT. Double-strand

break repair deficiency and radiation sensitivity inBRCA2 mutant cancer cells. J Natl Cancer Inst 1998;90:978–85.

8. Patel KJ, Yu VPCC, Lee H, et al. Involvement of Brca2in DNA repair. Mol Cell 1998;1:347–57.

9. Morimatsu M, Donoho G, Hasty P. Cells deletedfor Brca2 COOH terminus exhibit hypersensitivity tog-radiation and premature senescence. Cancer Res1998;58:3441–7.

10. Kraakman-van der Zwet M, Overkamp WJI, vanLange REE, et al. Brca2 (XRCC11) deficiency results inradioresistant DNA synthesis and a higher frequency ofspontaneous deletions. Mol Cell Biol 2002;22:669–79.

11. Hay T, Patrick T, Winton D, Sansom OJ, Clarke AR.Brca2 deficiency in the murine small intestine sensitizesto p53-dependent apoptosis and leads to the spontane-ous deletion of stem cells. Oncogene 2005;24:3842–6.

12. Bryant HE, Helleday T. Poly(ADP-ribose) polymeraseinhibitors as potential chemotherapeutic agents. Bio-chem Soc Trans 2004;32:959–61.

13. Bouchard VJ, Rouleau M, Poirier GG. PARP-1, adeterminant of cell survival in response to DNA damage.Exp Hematol 2003;31:446–54.

14. Farmer H, McCabe N, Lord CJ, et al. Targeting theDNA repair defect in BRCA mutant cells as atherapeutic strategy. Nature 2005;434:917–21.

15. Bryant HE, Schultz N, Thomas HD, et al. Specifickilling of BRCA2-deficient tumours with inhibitors ofpoly(ADP-ribose) polymerase. Nature 2005;434:913–7.

16. Miknyoczki SJ, Jones-Bolin S, Pritchard S, et al.Chemopotentiation of temozolomide, irinotecan, andcisplatin activity by CEP-6800, a poly(ADP-ribose)polymerase inhibitor. Mol Cancer Ther 2003;2:371–82.

17. Calabrese CR, Almassy R, Barton S. Anticancerchemosensitization and radiosensitization by the novelpoly(ADP-ribose) polymerase-1 inhibitor AG14361.J Natl Cancer Inst 2004;96:56–67.

18. Ireland H, Kemp R, Houghton C, et al. Inducible Cre-mediated control of gene expression in the murinegastrointestinal tract: effect of loss of h-catenin.Gasteroenterology 2004;126:1236–46.

19. Sansom OJ, Reed KR, Hayes AJ, et al. Loss of Apcin vivo immediately perturbs Wnt signaling, differenti-ation, and migration. Genes Dev 2004;18:1385–90.

20. Soriano P. Generalized lacZ expression with theROSA26 Cre reporter strain. Nat Genet 1999;21:70–1.

Research. on June 27, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

Brca2 Deficiency Sensitizes Cells to PARP Inhibition

In the article on how Brca2 deficiency sensitizes cells to PARPinhibition in the November 15, 2005 issue of Cancer Research (1),the grant support in the Acknowledgment should have read asfollows: Association for International Cancer Research grant no.03–336, KuDOS Pharmaceuticals Ltd., and Wales Gene Park.

1. Hay T, Jenkins H, Sansom OJ, Martin NMB, Smith GCM, Clarke AR. Efficientdeletion of normal Brca2 -deficient intestinal epithelium by poly(ADP-ribose)polymerase inhibition models potential prophylactic therapy. Cancer Res 2005;65:10145–8.

I2006 American Association for Cancer Research.doi:10.1158/0008-5472.CAN-66-1-COR

www.aacrjournals.org 599 Cancer Res 2006; 66: (1). January 1, 2006

Correction

2005;65:10145-10148. Cancer Res   Trevor Hay, Helen Jenkins, Owen J. Sansom, et al.   Models Potential Prophylactic TherapyEpithelium by Poly(ADP-Ribose) Polymerase Inhibition

-Deficient IntestinalBrca2Efficient Deletion of Normal

  Updated version

  http://cancerres.aacrjournals.org/content/65/22/10145

Access the most recent version of this article at:

   

   

  Cited articles

  http://cancerres.aacrjournals.org/content/65/22/10145.full#ref-list-1

This article cites 20 articles, 5 of which you can access for free at:

  Citing articles

  http://cancerres.aacrjournals.org/content/65/22/10145.full#related-urls

This article has been cited by 8 HighWire-hosted articles. Access the articles at:

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://cancerres.aacrjournals.org/content/65/22/10145To request permission to re-use all or part of this article, use this link

Research. on June 27, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from