[CANCER RESEARCH 58. 3965-3973, September I, 1998]

Mammalian 3-Methyladenine DNA Glycosylase Protects against the Toxicity andClastogenicity of Certain Chemotherapeutic DNA Cross-Linking Agents1

James M. Allan, Bevin P. Engelward,2 Andrew J. Dreslin, Michael D. VVvatt, Maria Tomas/., and Leona D. Samson3

Department of Cancer Cell Biology. Harvard School of Public Health. Boston. Massachusetts 02115 [J. M. A.. B. P.E.. A. J. D.. M.D.W.. L. D.S.I:Chemistry, Hunter College. City University of New York. New York. New York 10021 ¡M.T.]

nd Department of

ABSTRACT

DNA repair status is recognized as an important determinant of theclinical efficacy of cancer chemotherapy. To assess the role that a mammalian DNA glycosylase plays in modulating the toxicity and clastogenic-ity of the Chemotherapeutic DNA cross-linking alkylating agents, wecompared the sensitivity of wild-type murine cells to that of isogenic cellsbearing homozygous null mutations in the 3-methyladenine DNA glyco

sylase gene (Aag). We show that Aag protects against the toxic andclastogenic effects of l,3-bis(2-chloroethyl)-l-nitrosourea and mitomycin

C (MMC), as measured by cell killing, sister chromatid exchange, andchromosome aberrations. This protection is accompanied by suppressionof apoptosis and a slightly reduced p53 response. Our results identify3-methyladenine DNA glycosylase-initiated base excision repair as a po

tentially important determinant of the clinical efficacy and, possibly, thecarcinogenicity of these widely used Chemotherapeutic agents. However,Aag does not contribute significantly to protection against the toxic andclastogenic effects of several Chemotherapeutic nitrogen mustards (namely, mechlorethamine, melphalan, and chlorambucil), at least in the mouseembryonic stem cells used here. We also compare the Aag null phenotypewith the Fanconi anemia phenotype, a human disorder characterized bycellular hypersensitivity to DNA cross-linking agents, including MMC.Although Aag null cells are sensitive to MM( '-induced growth delay and

cell cycle arrest, their sensitivity is modest compared to that of Fanconianemia cells.

INTRODUCTION

Most antineoplastic Chemotherapeutic agents have significant mu-

tagenic and carcinogenic properties, and the incidence of some secondary cancers has been linked to high dose chemotherapy (1.2). Oneapproach to improving the clinical efficacy of Chemotherapeuticagents is to identify modulators of their activity that could potentiallybe exploited to sensitize target tumor tissue to therapy or to protectnontarget tissues during therapy. Such modulators include metabolicactivation pathways and DNA repair pathways.

It has become clear that cellular DNA repair status can be a majordeterminant of the clinical effectiveness of many Chemotherapeuticagents. For example, the mammalian MGMT4 protects against

BCNU-induced cell killing and SCE (3-6). Indeed, low MGMT

activity in bone marrow is thought to limit the clinical utility ofBCNU due to the severe myelosuppression induced by this agent

Received 3/23/98; accepted 6/24/98.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 with18 U.S.C. Section 1734 solely to indicate this fact.

' This work was supported by NIH Grants ROI CA55042 (to L. D. S.) and

ROICA28681 (to M. T.) and Individual Fellowship Award CA73135-OI (to M. D. W.)and by National Institutes of Environmental Health Training Grant 5T32ESO7I55 (toB. P. E.). J. M. A. is a Leukemia Society of America Fellow. B. P. E. is a PharmaceuticalManufacturers Association Foundation Advanced Predoctoral Fellow in Pharmacology/Toxicology, and L. D. S. is a Burroughs Wellcome Toxicology Scholar.

2 Present address: Division of Toxicology. Massachusetts Institute of Technology. 77

Massachusetts Avenue, Cambridge. MA 02139.1To whom requests for reprints should be addressed, at Department of Cancer Cell

Biology. Harvard School of Public Health. 665 Huntington Avenue. Boston. MA 02115.Phone: (617)432-1085: Fax: (617)432-0400: E-mail: lsamson@sph.harvard.edu.

4 The abbreviations used are: MGMT, O"-methylguanine DNA methyltransferase;

BCNU. 1.3-bis(2-chloroethyl)-l-nitrosourea: SCE. sister chromatid exchange; 3-MeA,3-methyladenine: BER. base excision repair: MMC. mitomycin C; FA. Fanconi anemia;

ES. embryonic stem; LIF. leukemia inhibitory factor: MMS. methyl methanesulfonate.

(7-9). The introduction of MGMT into bone marrow using retroviral

expression vectors and mice as a model successfully protects againstBCNU-induced toxicity (4, 8, 10-12). Conversely, inhibition of

MGMT activity has also been shown to sensitize cells and tissues toBCNU-induced toxicity and clastogenicity (13, 14). thus identifying

MGMT as a potential target for sensitizing tumor tissues.MGMT protection is mediated via a one-step reaction in which the

alkyl lesion on the O6 of guanine is transferred to an acceptor group

on MGMT, restoring the original DNA structure and inactivating therepair protein. Thus. MGMT repairs BCNU-induced O''-chloroethyl-

guanine lesions, which are precursors of DNA interstrand cross-links.

MGMT is virtually specific for the repair of alkyl lesions at theO6-guanine position in DNA (15-17), and so MGMT modulates the

toxicity of those chemotherapetic agents that attack DNA at thisposition, most notably the nitrosoureas, including BCNU [for reviewssee Lawley and Phillips (18) and Ludlum ( 19)]. In contrast, mammalian 3-MeA DNA glycosylase-initiated BER has a very broad substrate range (20). Both murine and human 3-MeA DNA glycosylases

(Aag and AAG, respectively, for alkyladenine DNA glycosylase)release not only 3-MeA, 3-methylguanine, and 7-methylguanine butalso 1.A^-ethenoadenine, hypoxanthine. and 7-hydro-8-oxoguanine(21-27). AAG also has in vitro base excision activity, albeit very lowcompared to that of 3-MeA, for DNA lesions induced by the Chemo

therapeutic nitrogen mustards chlorambucil, mechlorethamine, andmelphalan (28). Eukaryotic 3-MeA DNA glycosylases have also beenshown to protect against the cell-killing effects of Chemotherapeutic

nitrosoureas such as BCNU (29, 30) and of MMC (30). The formationof both intra- and interstrand DNA cross-links is thought to be

primarily responsible for the toxicity of nitrogen mustards, nitrosoureas, and MMC, although monoadducts may also contribute totheir toxicity. Here, we explore how 3-MeA DNA glycosylase activity

protects against the biological effects of these agents.We found previously that the murine 3-MeA DNA glycosylase Aag

protects against cell killing by MMC and BCNU (30). Here, we showthat this protection is mediated, at least in part, by the suppression ofapoptosis and that Aag-mediated DNA repair alters p53 induction bythese agents. Further. Aag protects against the chromosome-damaging

(clastogenic) effects of MMC and BCNU, as measured by SCE andchromosome aberration, identifying 3-MeA DNA glycosylase activity

as potentially important for suppressing the secondary tumors thatoften follow alkylating agent chemotherapy. However, Aag does notsignificantly protect against the Chemotherapeutic nitrogen mustards.Finally, we show that the MMC-sensitive phenotype of Aag null cells

is modest compared to that of FA cells.

MATERIALS AND METHODS

ES Cell Culture. ES cells were cultured on SNL76/7 mitotically inactivefeeder cells in DMEM. supplemented with 15% fetal bovine serum, 50units/ml penicillin, 50 /Mg/ml streptomycin. 2 mM glutamine. and 0.1 mM2-mercaptoethanol. Feederless ES cells were maintained in an undifferentiated

state in the presence of LIF [purified according to Mereau et al. (31)|.Drug Treatments. ES cells were treated with BCNU. MMC (Fluka, Mil

waukee, WI), mechlorethamine. melphalan. chlorambucil. hydrogen peroxide,tertiary-butyl hydrogen peroxide, or MMS (all from Sigma Chemical Co.. St.

3965

on March 16, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

DNA KIJ'AIR MOWLAÕÕ-S THE KKFFIT OF CHKMOTHERAPEUTICS

Louis, MO). For drug treatments, the cells were washed in I X PBS, andmedium was replaced with drug-containing serum-free ES medium supplemented with I,IF. Unless otherwise stated, drug treatments were for I h at 37°C

in 5% CO2, after which cells were washed in 1X PBS and reincubated in freshES medium supplemented with LIF.

Survival Curves. Aliquots of various ES cell dilutions were placed ontofeeder cell (SNL76/7 mitotically inactive)-coated 24-well plates. ES cells weretreated with drug the following day. After 5-9 days, dried colonies were

stained with Giemsa (Sigma) and counted. Killing curves were performed aminimum of three times, and Fig. 1 shows representative curves from eachtreatment.

SCEs and Chromosome Aberrations. For SCE and chromosome aberration measurements, 1 x 10" logarithmic-phase ES cells were seeded onto

gelatin-coated tissue culture dishes the day before drug treatment. Followingdrug treatment, cells were incubated in McCoy's 5A medium (Life Technol

ogies, Inc., Gaithersburg. MD) supplemented with LIF and 10 piM bromode-

oxyuridine for two rounds of DNA replication (24 h) for SCE analysis and forthe indicated limes for chromosome aberration analysis. Colcemid (0.2 /LIM)was included for the last 2 h of the bromodeoxyuridine treatment, and the cellswere subsequently harvested by mitotic shake-off, resuspended. and incubatedfor 15 min al 37°Cin hypotonie solution (0.2% potassium chloride. 0.2%sodium citrate, and 10% fetal bovine serum) and then fixed in Carnoy'ssolution (methanol-acetic acid. 3:1). To produce "harlequin" chromosome, a

modified fluorescence plus Giemsa technique was used (32). Slides werestained in Hoechst 33258 (5 fig per ml) for 20 min, mounted in 0.067 MSorensen's buffer (Na2HPO4-KH,PO4, pH 6.8) with a coverslip, and exposed

to a General Electric 15-W blacklight bulb (model F15T8, BLB) at 55°Cfor20 min. Slides were heated at 65°Cin 20x SSC for 20 min. rinsed, and stainedin a 5% Giemsa solution in Sorensen's phosphate buffer (0.067 M. pH 6.8). For

SCEs, 20 second division metaphase spreads were counted per data point. Forchromosome aberrations. 50 first-division metaphase spreads were counted per

data point.Chromosome aberrations were scored in a blinded study of metaphase

spreads. The frequency of chromosome breaks was estimated by weightedscoring of various types of aberrations according to Natajaran and Obe (33). Toaccurately reflect the quantity of aberrations within a population of cells, thenumber of break events was divided by the number of spreads examined.

Flow Cytometric Analysis of Cell Cycle Phase and Apoptosis. For flowcytometric analysis. 1 x l()6 logarithic-phase ES cells were seeded ontogelatin-coated 100-nim tissue culture dishes (or 0.5 X IO6 cells for 60-mm

dishes). Cells were treated with drug the following day and harvested post-

treatment as follows: medium was collected and retained, and cells weretrypsinized by the addition of 1.5 ml of 0.25% trypsin and incubated at 37°C

and 5% CO2 for 20 min. The trypsin was neutrali/ed by the addition of theretained medium. Cells were pelleted and fixed by resuspension in 1 ml of cold70% ethanol. Fixed cells (1 X IO6) were pelleted and resuspended in 5(X) /xl

of staining buffer (10 /ng/ml RNase A-40 /xg/ml propidium iodide, in IX

PBS). Cell samples were incubated at room temperature in staining solution forat least 1 h prior to analysis. Cells ( IO4) were analy/.ed for each sample using

a fluorescence-activated cell sorter (Ortho).

Assessment of Apoptosis. The apoptotic response was evaluated by threemethods. (</) Sub-G, cells were quantified by flow cytometry as described

previously (34). (b) A photometric ELISA was used for in vitro determinationof cytoplusmio histone-associated DNA fragments (Boehringer Mannheim.Indianapolis. IN). Briefly. 3 X IO6 cells were treated with genotoxic agent andharvested by trypsinization. and 3 X IO4 cells were processed according to themanufacturer's instructions, (c) Apoptotic nuclei were identified by micros

copy. Approximately IO4 cells fixed in 70% ethanol were dried onto micro

scope slides. Cells were stained in Hoechst 33258 buffer (0.5 /xg/ml Hoechst33258-0.05% nonfat dried milk) for 20 min at room temperature. Apoptotic

cells were identified by fluorescence microscopy. Five hundred cells wereexamined per slide.

Western Blot Analysis. Cells were boiled for 15 min in 1.7% SDS, 17%glycerol, 0.1 M DTT. 0.083 M Tris (pH 6.8), and 0.001% bromphenol blue ata density of 2.5 X IO4 cells/^l before loading. Samples (3 X 10' cell

equivalents per lane) were resolved on a 12% SDS-polyacrylamide gel. trans

ferred to a polyvinylidene difluoride membrane (Millipore. Bedford. Massachusetts), and probed with monoclonal p53 antibody (Pab 240; Santa CruzBiotechnology, La Jolla, California), followed by horseradish peroxidase-

conjugated rabbit antimouse IgG (Amersham. Arlington Heights. IL). Antibody binding was detected using enhanced chemiluminescence (Amersham).Equal loading was confirmed by Ponceau S staining of the membrane. Fig. 6Ashows representative blots.

Purification of Recombinant Aag. Mouse Aag cDNA was subclonedfrom pBEl. 1 (26) into the pCALn vector (Stratagene. Hercules, CA) to expressa 43 amino acid deleted Aag as an NH2-terminal fusion protein in-frame withthe M, 4000 calmodulin-binding protein. The pCALnAag vector was trans-fected into uikA iff#-deficient MV1932 DE3 Escherichia coli (a kind gift from

Dr. Anne Britt. University of California, Davis, CA). The pCALnAag containing E. coli were induced with 1 mM isopropyl-l-thio-ß-D-galactopyrano-side for 2 h at 37°Cand lysed by sonication, and then cell debris was removed

by centrifugation. The fusion protein was purified using a calmodulin affinityresin following the protocol specified by Stratagene. Protein isolated underidentical conditions as above from the same bacterial strain transfected withthe pCALn vector lacking the Aag cDNA provided the negative control.

Oligonucleotide-based Assay for DNA Glycosylase Activity. DNA oli-gonucleotides were 5'-end labeled with T4 polynucleotide kinase and

\y- "P]ATP. The MMC-cross-linked oligonucleotide was labeled on both

strands. All mono-oligonucleotides were labeled on the lesion containingstrand only. The unadducted oligonucleotide was 5'-labeled on the top strand

only (see Fig. 7,4). To test for DNA glycosylase activity. 1(X)fmol of labeleddouble-stranded oligonucleotide and 5.6 /j.g of Aag DNA glycosylase from the

purified Aag preparation were used. Equivalent protein concentrations wereused for the pCALn vector control incubations. The incubations were carriedout in buffer containing 20 mw Tris (pH 7.8), 100 mM potassium chloride, 2mM EDTA. 1 mM EGTA. and 5 mM 2-mercaptoethanol at 37°Cfor 120 min.

Following incubation, sodium hydroxide was added to 0.1 M, and the sampleswere heated to 70°Cfor 30 min to convert any abasic sites created by Aag into

DNA strand breaks. The DNA fragments were separated on a 20% denaturingpolyacrylamide gel. Gels were frozen and exposed to a phosphorimagingscreen (Bio-Rad, La Jolla. CA).

Growth Curves. Logarithmic-phase ES cells (IO4) were seeded per gela-

tini/ed 60-mm tissue culture dish. The following day, MMC (5-25 ng/ml) was

added to the medium. Cells were incubated for an additional 96 h and thentrypsinized and resuspended in ES medium; the cell yields per dish weremeasured by Coulter counter. Duplicate plates were measured. Cell growth indishes containing MMC was expressed as a percentage of the control untreateddish. Fig. 8,4 shows the mean and SD from three independent experiments.

RESULTS

Aag Protects against Cell Killing by Certain DNA Cross-Linking Agents. The cytotoxicity of chemotherapeutic agents is positively correlated with their antineoplastic activity. Similarly, themutagenicity and clastogenicity of chemotherapeutic agents is positively correlated with their carcinogenicity. Ideally, chemotherapeuticagents would be maximally cytotoxic for tumor cells, with minimalmutagenic and clastogenic side effects. To determine what role Aagplays in modulating cytotoxicity by DNA cross-linking agents, weexamined the sensitivity of Aag —¿�/—and Aag +/+ mouse ES cells to

the killing effects of MMC, BCNU, mechlorethamine, chlorambucil,and melphalan. in addition to other DNA-damaging agents. During

the course of these experiments, it was important to establish that thephenotype of the Aag —¿�/—and Aag +/+ ES cells was consistent withthat previously reported (30). As expected, Aag —¿�/—cells were

relatively sensitive to the killing effects of MMS, BCNU, and MMC(Fig. \,A-C: Ref. 30). In contrast, Aag —¿�/—and Aag +/+ cells were

equally sensitive to the killing effects of mechlorethamine, melphalan,and chlorambucil (Fig. 1, D-F). However, in some experiments, Aag-/- cells appeared mildly sensitive to the killing effects of chloram

bucil, but this phenotype was not consistently observed. We, therefore, conclude that Aag —¿�/—cells have a sensitivity to DNA cross-

linking agents that is not general but agent specific.Upon metabolic activation, MMC can react with DNA to form

monoadducts at the N2 and N7 positions of guanine (35-37). The N2

3966

on March 16, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

DNA REPAIR MODULATES THE EFFECT OF CHF.MOIÕII-.RAPEUTICS

(A)100

Fig. I. Cell killing in Aag -/- [clones 29 (D)

and 38 (O)] and Aag +/+ [clone 33 (•)]murine EScells in response to MMS (Ai. BCNU (B). MMC(C), mechloretharnine (D). melphatan (E). chloram-bucil (F). hydrogen peroxide (G). and tertiary-butyl

hydrogen peroxide (//). Cells were treated andplated onto feeder cell-coated 24-well dishes, andcolonies counted by Giemsa staining 5-9 days later.Each killing curve was performed at least threetimes, and representative plots are shown.

Mechloretharnine (uM)

(H)

0.01

Melphalan (uM)

O O O Q O(N ^ >O OO

Chlorambucil (uM)

So oo

Hydrogen peroxide(ug per ml)

lert-hydrogen peroxide

(Hgperml)

guanine monoadduct may subsequently rearrange to give either aDNA interstrand or intrastrand cross-link (38, 39). MMC can alsoundergo oxidation-reduction cycling to generate reactive oxygen species (40-43). Thus, in aerobic conditions, MMC may exert its toxicity

not only via direct interaction with DNA but also via the generation ofreactive oxygen species that, in turn, damage DNA. The mammalian3-MeA DNA glycosylases have been shown, in vitro, to release

oxidized guanine bases from DNA, making it possible that Aagprotects against MMC via the repair of oxidative DNA damage. Todetermine whether the sensitivity of the Aag —¿�/—cells to MMC may

be partly or wholly due to a reduced ability to repair oxidized DNAlesions, we tested their sensitivity to hydrogen peroxide and tertiary-butyl hydrogen peroxide. Aag —¿�/—and Aag +/+ cells were equally

sensitive to the cell-killing effects of both agents (Fig. l, G and H),suggesting that the MMC sensitivity of the Aag —¿�/—cells is due to

decreased repair of MMC-adducted DNA bases rather than decreased

repair of cytotoxic oxidized DNA bases.Aag Protects against the Chromosome-damaging Effects of

MMC and BCNU but not Those of Melphalan, Mechlorethamine,or Chlorambucil. To determine whether Aag protects against theclastogenic effects of the chemotherapeutic cross-linking agents, we

measured their ability to induce SCE and chromosome aberrations inAag +/+ versus Aag —¿�/—cells. We reported previously that Aag

protects against SCE induction by BCNU (Fig. 2A: Ref. 30). Here, wefind that Aag also provides significant protection against MMC-induced SCE because Aag —¿�/—cells were more sensitive than Aag

+/+ cells to SCE induction by MMC (Fig. 2B). However, these cellswere equally sensitive to SCE induction by mechloretharnine, mel-phalan, and Chlorambucil (Fig. 2, C-E).

Aag —¿�/—cells were also consistently more sensitive than Aag +/+

cells to chromosome aberrations induced by both MMC and BCNU(Figs. 3 and 4). When these aberrations are broken down into differenttypes, we see that the Aag —¿�/—cells are particularly sensitive to

chromosome-type aberrations (chromosome breaks and gaps) andonly modestly sensitive to chromatid-type aberrations (chromatid

breaks and gaps) induced by MMC and BCNU (Fig. 4, A and B).Taking the SCE and chromosome aberration data together, we inferthat Aag protects against the chromosome-damaging effects of BCNU

and MMC but does not contribute significantly to protection againstmechloretharnine, melphalan, or Chlorambucil.

Aag Suppresses Apoptosis Induction by MMC and BCNU. Theinduction of SCE and chromosome aberrations requires the transient formation of DNA strand break intermediates that couldpotentially signal the induction of apoptosis. Because Aag protectsagainst BCNU- and MMC-induced SCEs, chromosome aberrations, and cell killing (Figs. 1-4), we determined whether this

protection was accompanied by the suppression of apoptosis. Threedifferent markers of apoptosis were assayed: (a) chromatin condensation and fragmentation using Hoechst staining and microscopic examination; (b) the accumulation of DNA ends measuredby ELISA; and (c) the accumulation of cells with a sub-G, DNAcontent measured by flow cytometry (see "Materials and Methods"). Aag —¿�/—cells were consistently more sensitive than Aag

+/+ cells to BCNU- and MMC-induced apoptosis (Fig. 5). Wedemonstrated previously that Aag-proficient and -deficient cells

are equally sensitive to apoptosis induced by ionizing radiation,indicating that Aag —¿�/—cells are not generally more sensitive toapoptosis in response to DNA damage (44). We infer that MMC-and BCNU-induced base lesions that are inefficiently repaired inAag —¿�/—cells can elicit apoptosis.

3967

on March 16, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

DNA REPAIR MODULATES THE EFFECT OF CHEMOTHERAPEUTICS

(A) (B)

2.5-, 1 1.25

2-

uSii-pO

W

1/20.5-

Fig. 2. SCE in Aag -/- (Q) and Aag +/+ (•)murine ES

cells in response lo BCNU (A), MMC (B). chlorambucil (O.mechloriMhaniine (/)), and melphahn (E). A and H. tlitui points.means from three independent experiments; hars. SD. C-E,until fioint\. means from two independent experiments.

BCNU (UM)

(C) (D)

MMC (ng per ml)

(E)

Chlorambucil (u,M)Mechlorethamine (u,M)

Melphalan(uM)

Apoptosis can be mediated via p53-dependent and -independent

pathways (45), but whether the apoptosis that Aag suppresses involvesp53 is not yet clear. To begin to address this question, we monitoredMMC and BCNU induction of p53 in Aag +/+ versus Aag —¿�/-cells.p53 accumulation was maximal in both Aag —¿�/—and Aag +/+ cells

at 24-36 h posttreatment with either MMC or BCNU. For both

agents, we consistently observed a slightly greater accumulation ofp53 in the Aag -/— cells versux the Aag +/+ cells (Fig. 6). We

previously demonstrated that ioni/ing radiation induces similaramounts of p53 in Aag —¿�/—and Aag +/+ cells (44), indicating thatthe Aag —¿�/—cells do not have a generalized increase in their p53induction response to DNA damage. We infer that MMC- and BCNU-induced base lesions that are inefficiently repaired in Aag —¿�I—cells

can signal p53 induction, albeit weakly.Aag Does Not Have in Vitro Excision Activity for the MMC

/V2-Guanine-;V2-Guanine Interstrand Cross-Link or ¿V2-Guanine

Monoadduct. It seems reasonable to assume that Aag provides protection against MMC by initiating the repair of DNA base lesionsinduced by this agent. However, exactly which MMC-induced base

lesions Aag recognizes is not yet known. To begin to investigate this,we determined whether Aag could act on two major MMC-inducedDNA lesions, namely, the A^-guanine monoadduct and the iV2-gua-nine-/V2-guanine interstrand cross-link. Double-stranded oligonucleo-

tides containing these lesions at specific sites (Fig. 1A) were preparedas described previously (46) and used as substrates for the purified

Aag enzyme. However, Aag had no in vitra excision activity for eitherthe MMC A^-guanine monoadduct or the MMC ¿V2-guanine-¿V2-gua-

nine interstrand cross-link (Fig. IB, Lanes 6 and 8, respectively). Theinterstand cross-link runs as a 43-mer under denaturing conditions(Fig. IB, Lanes 7 and 8): this would have been converted to a 9- anda 23-mer (which would appear as a 24-mer due to the MMC lesion)had Aag excised the MMC-adducted guanine on the top strand but not

the bottom strand (Fig. 1A). Alternatively, had Aag excised theMMC-adducted guanine on the bottom strand alone then a 14- and a20-mer (which would have appeared as a 21-mer due to the MMClesion) would have been generated (Fig. 1A). Excision of both MMC-adducted guanines would have generated a 9- and a 14-mer (Fig. 1A).

Under denaturing conditions, the MMC-monoadducted oligonu-cleotide runs as 20-mer (Fig. IB, Lanes 5 and 6). Excision of theMMC A^-guanine monoadduct by Aag would have converted the

20-mer to a 9-mer (Fig. 7/4).

As a control for these experiments, we show that purified Aag hasin vitro excision activity for an oligonucleotide site-specifically modified with hypoxanthine (Fig. IB, Lane 2). The hypoxanthine-contain-ing 25-mer is converted to a 12-mer by the action of Aag and sodium

hydroxide.The Aag Null Phenotype Confers a Modest Sensitivity to MMC-

induced Growth Delay and G2-M Cell Cycle Arrest. FA is an

autosomal recessive human disorder that is characterized by certaindevelopmental abnormalities, bone marrow failure, and increased

3968

on March 16, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

DNA RKI'AIR MODULATES THE EFFECT OF CHEMOTHERAPEUTICS

(A)

1.25BCNU14h

(B)50 UM BCNU

- 0.8-

I 0.6 H

0.4-

0.2v~i o u"i »n >n o^ - ¡N - ¿ M

Hrs post 50uM BCNU

(D)

1.250.3 ng per ml MMC

1-

SL 0.75-

0.5-

0.25-

MMC (|j.g per ml) Hrs post 0.3|ig per ml MMC

Fig. 3. Chromosome aberration induction in Aag -/— (D) and Aag +/+ (•)murineES cells in response (o BCNU (A and B) and MMC (C and D). 14-16 h posttrealment (Aand O and over time (fi and £>).

cancer susceptibility. Aag —¿�/—cells share with FA cells the distinc

tive phenotype of MMC sensitivity (47). However, the MMC sensitivity of FA cells has been traditionally measured using differentparameters from those thus far discussed for Aag —¿�/—cells, making

the sensitivity of these two cell types difficult to compare. Specifically, FA cells have been characterized by a pronounced MMC-induced G2-M arrest and a dramatic growth inhibition in the contin

uous presence of MMC, as measured by simple monitoring of cellnumbers (48-51). To determine whether the MMC sensitivity of Aag—¿�/—cells was comparable to that reported for FA cells, we applied

the same measures of sensitivity. Fig. 8.4 displays the modest growthinhibition of Aag —¿�/—cells versus Aag +/+ cells after continuous

MMC exposure. Cell numbers were measured after a growth periodequivalent to that used for measuring FA cell sensitivity. The MMCconcentration that produces 50% growth inhibition (ICMI)for FA cellsis 5-30-fold lower than that for normal cells, depending on the cellline tested (50, 51 ). In contrast, the MMC IC50 for Aag —¿�/—cells was

only marginally lower than that for Aag +/+ cells (Fig. 8.4). Inaddition, the MMC-induced G2-M delay is very modest for Aag —¿�/—

versus Aag +/+ cells and is only transiently discernible at 24 and 32 hposttreatment (Fig. 8fi). In contrast, FA cells display a robust and veryprolonged MMC-induced G2-M delay (48,49). Thus, although the FAand Aag —¿�/—phenotypes share certain similarities, the severity of theMMC-sensitive phenotype is much less pronounced in Aag —¿�/—cells,

suggesting that the phenotypes arise from distinct cellular defects.

DISCUSSION

Bifunctional alkylating agents that can form DNA cross-links are

commonly used for cancer chemotherapy. Although these agentsinduce many different types of DNA lesions, the intrastrand andinterstrand cross-links are considered critical for their antineoplasticactivity. The formation and repair of DNA cross-links may. therefore,

determine the clinical activity of the bifunctional alkylating agents. Inthis study, we explored the role that the Aag 3-MeA DNA glycosylase

repair enzyme plays in modulating the biological activity of severalclinically important bifunctional alkylating agents. One approach toexamining the role of a DNA repair enzyme in protecting against aDNA-damaging agent is to determine how enzyme-deficient mutants

respond to that agent. Our results suggest that BER initiated by theAag enzyme confers significant protection against the cytotoxic andclastogenic effects of both BCNU and MMC and that this protectionis accompanied by decreased apoptosis. In contrast. Aag does notappear to protect against the chemotherapeutic nitrogen mustardsmelphalan. chlorambucil. and mechlorethamine.

(A)Mitomycin C

(B)BCNU

0.3

0.25-

0.2-

0.15-

0.1-

0.05-

Fig. 4. Qualitative analysis of chromosome aberrations induced in Atta +/+ (•)andAag -/- (D) murine ES cells in response to 0.05-0.3 jig/ml MMC (Ai and 25-75 fiM

BCNU (fì).Four hundred melaphase spreads were analyzed for each treatment, and equalnumbers of spreads from various doses were used for Aag -/- and Aag +/+. The

following aberrations were monitored: chromatid breaks and gaps (TB&C). chromosomebreaks and gaps (SB&G). triradial chromosomes (Tri), quadriradial chromosomes (Qitiul).dicentric chromosomes (Dicen), and chromosome deletions (Del).

3969

on March 16, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

DNA REPAIR MODULATES THE EFFECT OF CHEiMOI HI-.RAPEUTICS

Fig. 5. Apoplosis in Aag —¿�/—(D) and Aag +/+(•)murine ES cells in response Io BCNU (A-C) andMMC (D-F), as measured by Hoechst analysis overlime (A and D). by ELISA over lime (B and £).and byflow cytomelry (C and F}. C and F, samples weretaken at 36 and 48 h posltrealmenl. respectively. Datapointa, means from three to five independent experiments; bars, SE (noie that bars in A and D are toosmall to be discernible).

I

(B)ELISA Flow Cytometry

Hrs post 250U.M BCNU

' ' Hoechst

Hrs post l (ig per ml MMC

O 0 S ? 3

Hrs post lug per ml MMCMMC (ng per ml)

l.SngpermlMMC

250nM BCNU

Hrs post MMC Hrs post BCNU

Fig. 6. A. p53 protein levels in Adg —¿�/—and Atig +/+ murine ES cells in response to

250 /AMBCNU and 1.5 ¿ig/mlMMC over time. This experiment was performed fivetimes, and representative blots are shown. B. quantitative analysis of p53 protein inductionin Aitg —¿�/—(O) and Aag +/-t- (•)murine ES cells in response to 1.5 ^tg/nil MMC and

175 fiM BCNU. Dula points, mean protein inductions from three independent experiments; fairs. SD. Optical density was measured using Gel Doc apparatus and MolecularAnalyst software (both from Bio-Rad) using volume analysis with subtraction of local

background.

Aag presumably provides protection against MMC and BCNU byinitiating the repair of DNA base lesions induced by these agents.MMC induces monoadducts at the N2 and N1 positions of guanine(35-37); the N2 monoadduct can go on to form an /V2-guanine-/V2-

guanine interstrand or intrastrand cross-link (38. 39). Our finding that

Aag has no apparent in vitro excision activity for either of theselesions does not eliminate them as possible in vivo substrates for Aagbecause our in vitro assay may not be sensitive enough to detect aminor activity. Alternatively, Aag may have activity against otherMMC-induced DNA base lesions, namely, the A^-guanine monoadduct or the A^-guanine-A^-guanine intrastrand cross-link or some

other as yet uncharacterized MMC-induced base lesions.Although MMC is known to create toxic interstrand cross-links, the

ability of this agent to promote oxidative DNA damage is also thoughtto contribute to its biological effects. Indeed, because Aag has in vitroactivity against oxidized guanines (7-hydro-8-oxoguanine)5 (21), it

seemed feasible for Aag to protect against MMC via the repair ofoxidative DNA damage. However, Aag —¿�/—and Aag +/+ cells are

equally sensitive to the killing effects of hydrogen peroxide andtertiary-butyl hydrogen peroxide, and from this, we infer that Aag

does not contribute significantly to the repair of cytotoxic oxidativeDNA damage in vivo. This result is consistent with the observationthat Aag has no effect on apoptosis, p53 induction, or cell cycle arrestinduced by -y-radiation (44). an agent known to generate significant

levels of oxidative DNA damage (52).BCNU induces lesions at several sites on DNA, predominantly at

the N7 position of guanine but also at the Ob position of guanine

5 M. D. Wyatt and L. D. Samson, unpublished observations.

3970

on March 16, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

DNA REPAIR MODULATES THE EFFECT OF CHF.MOÕHF.RAPËUTICS

(A)TTorii .Hnr„ 5 '-AATACATCCGACTTACTCAAUnadducted Oligo 3 , .TTATGTAGGCTGAATGAGTTAGG

Monoadducted Oligo

5'-AATACATCCGACTTACTCAAMK'

3'-TTATGTAGGCTGAATGAGTTAGG

5'-AATACATCCGACTTACTCAACross-linkedOligo MMC'

3'-TTATGTAGGCTGAATGAGTTAGG

H*nuâ„¢5'-GGATAGTGTCCAHxGTTACTCGAAGCMXuugo3,_CCTATCACAGGTTCAATGAGCTTCG

(B)

Hxun- mono- cross-

adducted adduct link

I I<U CO

^ <43-mer25-mer20-mer

12-mer

—¿�*—¿�—>_>

123456789Fig. 7. /H n/m DNA giycosylase excision activity by Aag for unadductcd oligonu-

cleotide (Lane 4i and hypoxanthine (Lane 2), MMC /V2-guanine (Lune 6}. and MMCA/2-guanine-/V2-guanineinterstrand cross-link (Ijinc 8) site specifically modified oligo-nucleotides (see "Materials and Methods"). Also included are control protein extracts for

the hypoxanthine and unadducted. MMC-monoadducted. and MMC-cross-linked substrates (Lanes I. j. 5, and 7. respectively).

[reviewed by Ludlurn (19)]. If the 06-chloroethylguanine lesion es

capes repair by MGMT, it can go on to rearrange into a \,Cf-

ethanoguanine lesion, thought to be the immediate precursor of themajor class of BCNU-induced interstrand cross-links (53). The similarity of the 1,O''-ethanoguanine lesion and KA^-ethenoadenine

(which Aag excises efficiently, as does the homologous human enzyme: Refs. 22, 24, and 27) leads us to speculate that at least one wayfor Aag to provide BCNU resistance may be via the repair of 1,O6-

ethanoguanine lesions. Such repair would prevent DNA interstrandcross-link formation. Alternatively, Aag may protect against BCNU-induced toxicity via the repair of A/7-guanine or other lesions. Indeed,

yeast and bacterial 3-MeA DNA glycosylases have in vitro excisionactivity for both the A/7-chloroethylguanine and A/7-hydroxyethylgua-

nine base lesions (19, 29, 54, 55). Whether the mammalian enzymesdisplay similar activities is currently being studied.

The human and rat 3-MeA DNA glycosylases are known to have in

vitro activity against DNA base lesions induced by chlorambucil,mechlorethamine, and melphalan (28). However, we find that Aagnull cells are not significantly sensitive to killing and clastogenesis byany of these agents. These observations need not be contradictory, andcould be explained in any of the following ways, (a) The mouseenzyme may simply not have activity for chlorambucil-, mechlorethamine-, or melphalan-induced lesions (this has yet to be tested), (b)

The in vivo excision activity of the mouse enzyme (and perhaps thehuman and rat enzymes) may not be efficient enough to have biological consequences, (r) The lesions excised may not be cytotoxic orclastogenic. (d) Other DNA repair pathways (or DNA glycosylases)may be able to compensate for the decreased BER in an Aag nullbackground. Indeed, in E. coli, nucleotide excision repair is the majorrepair pathway for mechlorethamine-induced DNA lesions, and BERcontributes relatively little to protecting cells from mechlorethamine-

induced cell killing (28): this may also be true for mouse ES cells.Although the Aag null and FA phenotypes are similar in that both

show MMC-induced G2-M phase cell cycle arrest and growth delay,

they differ in that the Aag null cells are only modestly more sensitivethan their wild-type counterparts. Unlike FA cells, Aag null cells do

not show spontaneous growth delay, cell cycle arrest or elevatedchromosome breakage (30, 48, 49, 56, 57). Aag —¿�/—cells are particularly sensitive to chromosome-type aberrations in response to

(A)

O v> o >n—¿�—¿�fs tN

MMC (ng per ml)

24 h

Relative Fluorescence

32 h

Fig. 8. (A) Growth of Aag —¿�/—(O) and Aag +/+ (•)murine ES cells in continuous exposure to MMC. Cell growth was assessedby Coulter counter analysis 4 days post-MMCaddition. Dala /minis, means from three independent experiments: bars. SD. B. flow cytometric analysis of Aag -/- and Aag +/+ murine ES cells post-MMC treatment. Cells were

treated for l h with 2 /Ag/rnl MMC. and cell cycle analysis by DNA content was assessedby flow cytometry up to 32 h posttreatment. Ten thousand cells were analyzed per sample,and cells with sub-G, DNA content arc included in the analysis. This experiment was performed five times, and a representative example is shown.

3971

on March 16, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

DNA REPAIR MODULATES THE EFFECT OF CHF.MOTHERAPEUTICS

MMC and are not very sensitive to chromatid-type aberrations in

duced by this agent. In contrast, FA cells are predominantly sensitiveto MMC-induced chromatid-type aberrations (57). This result suggests that MMC-induced aberrations in Aag null and FA cells origi

nate from different primary lesions or that different mechanisms ofaberration-induction are operating.

In summary, we have shown that the murine 3-MeA DNA glyco-

sylase, Aag, protects against cytotoxicity and clastogenicity inducedby the chemotherapeutic cross-linking agents MMC and BCNU but

not against those induced by the nitrogen mustards mechlorethamine,melphalan, and chlorambucil. It remains to be determined if thehomologous human enzyme, AAG, also protects against MMC andBCNU. If so, then AAG may represent an important modulator ofchemotherapeutic efficacy.

ACKNOWLEDGMENTS

1. Kaldor. J. M., Day. N. E.. Band. P.. Choi. N. W., Clarke. E. A.. Coleman. M. P..Hakama. M., Koch, M.. Langmark, F.. Neal, F. E., Pettersson, F.. Pompe-Kim. V..Prior. P.. and Slorm. H. H. Second malignancies following testicular cancer, ovariancancer, and Hodgkin's disease: an international collaborative study among cancer

registries. Int. J. Cancer. 39: 571-585. 1987.2. Swerdlow, A. J.. Douglas. A. J.. and Hudson, C. V. Risk of second primary cancers

after Hodgkin's disease by type of treatment: analysis of 2846 patients in the British

National Lymphoma Investigation: relationships to host factors, histology and stageof Hodgkin's disease, and splenectomy. Br. Med. J., 304: 1137-1143, 1992.

3. Erickson, L. C.. Laurent. G.. Sharkey, N. A., and Kohn. K. W. DNA cross-linking andmonoadduct repair in nitrosourea-treated human tumour cells. Nature (Lond.). 288:727-729. 1980.

4. Preuss. I.. Thus!. R., and Kaina, B. Protective effect of 06-methylguanine-DNA

methyllransferase (MGMT) on the cytotoxic and recombinogenic activity of differentantineoplastic drugs. Int. J. Cancer. 65: 506-512. 1996.

5. Schwartz, J. L.. Turkula. T.. Sagher. D., and Strauss. B. The relationship betweenf'-alkylguanine alkyllransfera.se activity and sensitivity to alkylation-induced sister

chromatid exchanges in human lymphoblastoid cells. Carcinogenesis (Lond.), 10:681-685, 1989.

6. Bodell. W. J.. Aida, T.. Berger, M. S., and Rosenblum. M. L. Increased repair ofO^-alkylguanine DNA adducts in glioma-derived human cells that are resistant to the

cytotoxic and cytogenetic effects of 1.3-bis(2-chloroethyl)-1 -nitrosourea. Carcinogenesis (Lond.). 7: 879-883. 1986.

7. Gerson, S. L., Trey. J. E.. Miller. K., and Berger, N. A. Comparison of O6-

alkylguanine-DNA alkyltransferase activity based on cellular DNA content in human,rat and mouse tissue. Carcinogenesis (Lond.I. 7: 745-749, 1986.

8. Moritz. T.. Mackay. W., Glassner. B. J.. Williams. D. A., and Samson, L. Retrovirus-mediatcd expression of a DNA repair protein in bone marrow protects hematopoieticcells from nitrosourea-induced toxicity in vitra and in vivo. Cancer Res., 55: 2608-

2614. 1995.9. Schabe). F. M.. Jr. Nitrosoureas: a review of experimental antitumour activity. Cancer

Treat. Rep.. 60: 665-698. 1976.

10. Maze, R., Carney. J. P., Kelley. M. R.. Glassner, B. J.. Williams, D. A., and Samson.L. D. Increasing DNA repair methyltransferase levels via bone marrow stem celltransduclion rescues mice from the toxic effects of 1.3-bis(2-chloroethyl)-l-nitrosourea. a chemotherapeutic alkylating agent. Proc. Nati. Acad. Sci. USA, 93: 206-210. 1996.

11. Davis, B. M., Reese. R. J.. Koc, O. N.. Lee, K., Schupp, J. E.. and Gerson, S. L.Selection for G156A O-6-methylguanine DNA methyltransferase gene-induced hematopoietic progenitors and protection from lethality in mice treated with 0-6-benzylguanine and 1.3-bis(2-chloroethyl)-l-nilrosourea. Cancer Res., 57: 5093-5099.

1997.12. Wang, G., Weiss, C.. Sheng, P. H.. and Bresnick, E. Retrovirus-mediated transfer of

the human O-6-melhylguanine-DNA methyltransferase gene into a murine hemato

poietic stem cell line and resistance to the toxic effects of certain alkylating agents.Biochem. Pharmacol.. 51: 1221-1228. 1996.

13. Chinnasamy, N.. Rafferty, J. A., Hickson, I.. Ashby, J., Tinwell. H.. Margison, G. P.,Dexter, T. M.. and Fairburn. L. J. Oft-benzylguanine potentiates the in vivo toxicity

and clastogenicity of temozolomide and BCNU in mouse bone marrow. Blood. 89:1566-1573. 1997.

14. Dolan. M. E., Mitchell. R. B.. Mummen. C., Moschel. R. C.. and Pegg. A. E. Effectof O6-benzylguanine analogues on sensitivity of human tumor cells to the cytotoxiceffects of alkylating agents. Cancer Res.. 51: 3367-3372. 1991.

15. Zak, P., Kleibl. K.. and Laval. F. Repair of O6-methylguanine and O4-methylthymineby the human and rat O6-methylguanine-DNA methyltransferases. J. Biol. Chem..

269: 730-733, 1994.

16.

18.

19.

20.

21.

22.

23.

24.We are grateful for the expert technical assistance of Amy Imrich, HarvardSchool of Public Health, supported by Environmental Health Sciences CenterGrant 5P30 ES0002. 25

REFERENCES

26.

27.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

Sassanfar. M.. Dosanjh. M. K.. Essigmann, J. M.. and Samson. L. Relative efficiencies of the bacterial, yeast, and human DNA methyllransferases for the repair ofO6-methylguanine and O4-methylthymine. J. Biol. Chem.. 266: 2767-2771, 1991.

Samson, L., Han, S., Marquis, J. C., and Rasmussen, L. J. Mammalian DNA repairmethyltransferases shield O* MeT from nucleotide excision repair. Carcinogenesis

(Lond.). 18: 919-924, 1997.

Lawley, P. D.. and Phillips. D. H. DNA adducts from chemotherapeutic agents.Mutât.Res., 355: 13-40, 1996.

Ludlum. D. B. DNA alkylation by the haloethylnitrosoureas: nature of modificationsproduced and their enzymatic repair or removal. Mutât.Res., 233: 117-126, 1990.

Singer. B.. and Hang. B. What structural features determine repair enzyme specificityand mechanism in chemically modified DNA. Chem. Res. Toxicol.. 10: 713-732.

1997.Bessho. T., Roy. R.. Yamamoto, K., Kasai, H.. Nishimura. S., Taño,K., and Mitra,S. Repair of 8-hydroxyguanine in DNA by mammalian A'-methylpurine-DNA glyco-sylase. Proc. Nati. Acad. Sci. USA, 90: 8901-8904. 1993.Dosanjh. M. K.. Roy. R.. Mitra, S., and Singer. B. 1.A'o-ethenoadenine is preferredover 3-methyladenine as a substrate by a cloned human A'-methylpurine-DNA gly-cosylase (3-methyladenine-DNA glycosylase). Biochemistry, 33: 1624-1628. 1994.

Saparbaev. M.. and Laval. J. Excision of hypoxanthine from DNA containing dIMPresidues by the Escherichia coli, yeast, rat. and human alkylpurine DNA glycosylase.Proc. Nati. Acad. Sci. USA. 91: 5873-5877, 1994.

Saparbaev, M.. Kleibl. K.. and Laval, J. Escherichia coli, Saccharomyces cerevlsìae,rat and human 3-methyladenine DNA glycosylases repair 1, /V6-ethenoadenine whenpresent in DNA. Nucleic Acids Res., 23: 3750-3755, 1995.Singer. B.. Antoccia. A., Basu. A. K., Dosanjh. M. K., Fraenkel-Conrat, H.,Gallagher, P. E., Kusmierek, J. T., Qui, Z-H.. and Rydberg, B. Both purified human l-/V6-ethenoadenine-binding protein and purified human 3-methyladenine-DNAglycosylase act on l.A'o-ethenoadenine and 3-methyladenine. Proc. Nati. Acad.Sci. USA, 89: 9386-9390. 1992.

Engelward. B. P., Boosalis. M. S., Chen, B. J.. Deng. Z., Siciliano. M. J.. and Samson,L. D. Cloning and characterization of a mouse 3-methyladenine/7-methylguanine/3-

methylguanine DNA glycosylase cDNA whose gene maps to chromosome 11. Carcinogenesis (Lond.). 14: 175-181. 1993.

Engelward. B. P.. Weeda. G., Wyatt. M. D., Broekhof. J. L. M., de Wit, J., Donker.!.. Allan, J. M., Gold, B., Hoeijmakers, J. H. J.. and Samson, L. D. Base excisionrepair deficient mice lacking the Aag alkyladenine DNA glycosylase. Proc. Nail.Acad. Sci. USA. 94: 13087-13092, 1997.Mattes. W. B.. Lee. C-S., Laval, J., and O'Connor, T. R. Excision of DNA adducls

of nitrogen mustards by bacterial and mammalian 3-methyladenine DNA glycosylases. Carcinogenesis (Lond.), 17: 643-648, 1996.

Matijasivec. Z., Boosalis. M., Mackay, W., Samson. L., and Ludlum, D. B. Protectionagainst chloroethylnitrosourea cytotoxicity by eukaryotic 3-methyladenine DNA glycosylase. Proc. Nati. Acad. Sci. USA. 90: 11855-11859, 1993.

Engelward. B. P.. Dreslin. A.. Christensen. J.. Huszar, D.. Kurahara. C., and Samson,L. Repair-deficient 3-methyladenine DNA glycosylase homozygous mutant mousecells have increased sensitivity to alkylation-induced chromosome damage and cellkilling. EMBO J.. 15: 945-952, 1996.Mereau, A., Grey. L.. Piquet-Pellorce. C.. and Heath. J. K. Characterization of a

binding protein for leukemia inhibitory factor localized in extracellular matrix. J. Cell.Biol., ¡22:713-719, 1993.

Perry. P., and Wolff, S. New Giemsa method for the differential staining of sisterchromatids. Nature (Lond.), 25/: 156-158, 1974.

Natajaran. A. T.. and Obe. G. Molecular mechanisms involved in the production ofchromosomal aberrations. III. Restriction endonucleases. Chromosoma. 90: 120-127,

1984.Nicoletti, I., Migliorati. G., Pagliacci. M. C., Grignami, F., and Riccardi, C. A rapidand simple method for measuring thymocyte apoptosis by propidium iodide stainingand flow cytometry. J. Immunol. Methods. 139: 271-279, 1991.

Tomasz. M.. Chawla, A. K.. and Lipman, R. Mechanism of monofunctional andbifunclional alkylation of DNA by mitomycin C. Biochemistry. 27: 3182-3187, 1988.

Tomasz. M.. Lipman, R., McGuinness, B. F., and Nakanishi. K. Isolation andcharacterization of a major adduci between mitomycin C and DNA. J. Am. Chem.Soc., 110: 5892-5896. 1988.Kumar. G. S.. Lipman. R.. Cummings. J.. and Tomasz, M. Mitomycin C-DNAadducts generated by DT-diaphorase: revised mechanism of the enzymatic reductiveactivation of mitomycin C. Biochemistry, 36: I4128-I4I36, 1997.

Tomasz, M., Lipman. R.. Chowdary. D.. Pawlak, J.. Verdine, G. L., and Nakanishi,K. Isolation and structure of a covalent cross-link adduci between milomycin C andDNA. Science (Washington DC), 235: 1204-1208. 1987.

Bizanek. R.. McGuinness, B. F., Nakanishi, K., and Tomasz, M. Isolation andstructure of an intrastrand cross-link adduci of mitomycin C and DNA. Biochemistry,31: 3084-3091. 1992.

Bachur. N. R.. Gordon. S. L.. and Gee. M. V. A general mechanism for microsomalactivation of quinone anticancer agents to free radicals. Cancer Res., 38: 1745-1750,

1978.Bachur. N. R.. Gordon. S. L.. Gee, M. V.. and Kon. H. NADPH cytochrome P-450reducíaseactivation of quinone anticancer agents to free radicals. Proc. Nati. Acad.Sci. USA, 76: 954-957, 1979.

Pristos. C. A., and Sartorelli. A. C. Generation of reactive oxygen radicals throughbioactivation of mitomycin antibiotics. Cancer Res., 46: 3528-3532. 1986.Tomasz. M. H2O3 generation during the redox cycle of mitomycin C and DNA-boundmitomycin C. Chem. Biol. Interact.. 13: 89-97, 1976.

3972

on March 16, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

DNA REPAIR MODULATES THE EFFECT OF CHEMOTHERAPEUTICS

44. Engelward. B. P.. Allan. J. M.. Dreslin, A. J., Kelly, J. D.. Wu, M.. Gold, B.. andSamson. L. D. A chemical and genelic approach together define the biologicalconsequences of 3-methyladenine lesions in the mammalian genome. J. Biol. Chem..273: 5412-5418, 1998.

45. Clarke, P. R., Purdie. C. A., Harrison, D. J.. Morris, R. G., Bird, C. C., Hooper, M. L.,and Wyllie. A. H. Thymocyte apoptosis induced by p53-dependenl and independentpathways. Nature (Lund.). 362: 849-852. 1993.

46. Kumar, G. S., Johnson. W. S., and Tomasz, M. Orientation isomers of the mitomycin Cinierstrand cross-link in non-self-complementary DNA. Biochemistry, 32: 1364-1372,

1993.47. Sasaki. M. S., and Tonomura, A. A high susceptibility of Fanconi's anemia to chromo

some breakage by DNA cross-linking agents. Cancer Res., 33: 1829-1836, 1973.48. I inn ill.m v B., Aurias, A.. Dutrillaux. A. M.. Buriot, D., and Prieur. M. The cell cycle

of lymphocytes in Panconi anemia. Hum. Genet.. 62: 327-332. 1982.49. Kubbies. M.. Schindler, D.. Hoehm, H.. Schinzel. A., and Rabinovitch. P. S. Endog

enous blockage and delay of the chromosome cycle despite normal recruitment andgrowth phase explain poor proliferation and frequent endomitosis in Panconi anemiacells. Am. J. Hum. Genet.. 37: 1022-1030, 1985.

50. Kruyt, P. A. E.. Dijkmans, L. M., van den Berg. T. K.. and Joenje. H. Panconi anemiagenes act to suppress a cross-linker-inducible p53-independent apoptosis pathway inlymphoblastoid cell lines. Blood. 87: 938-948, 1996.

51. Ridet, A., Guillouf. C.. Duchaud. E., Cundan, E.. Fiore, M.. Moustacchi. E., andRossetti. F. Deregulated apoplosis is a hallmark of the Panconi anemia syndrome.Cancer Res., 57: 1722-1730, 1997.

52. Dizdaroglu, M. Oxidative damage to DNA in mammalian chromatin. Mutât.Res.,275: 331-342.

53. Tong, W. P., Kirk. M. C., and Ludlum. D. B. Formation of the cross-link. \-lN*-deoxycytidyl]-2-[A'l-deoxyguanosyl]-ethane. in DNA treated with. N. /V'-bis-(2-

chloroethyl)-A'-nitrosourea (BCNU). Cancer Res.. 42: 3102-3105, 1982.54. Carter, C. A., Habraken, Y.. and Ludlum. D. B. Release of 7-alkylguanines from

haloethylnitrosourea-treated DNA by E. coli 3-methyladenine-DNA glycosylase II.Biochem. Biophys. Res. Commun., 155: 1261-1265.

55. Habraken, Y.. Carter, C. A., Kirk. M. C.. and Ludlum. D. B. Release of 7-alkylguanines from W-(2-chloroethyl)-Ar-cyclohexyl-A/-nitrosourea-modified DNA by3-methyladenine DNA glycosylase II. Cancer Res., 51: 499-503. 1991.

56. Weksberg. R., Buchwald. M.. Sargent, P.. Thompson. M. W.. and Siminovitch. L.Specific cellular defects in patients with Panconi anemia. J. Cell Physiol.. 101:311-324, 1979.

57. Miura. K., Morimolo. K., and Koizumi. A. Proliferalive kinetics and mitomycinC-induced chromosome damage in Fanconi's anemia lymphocytes. Hum. Genet., 63:

19-23, 1983.

3973

on March 16, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from

1998;58:3965-3973. Cancer Res   James M. Allan, Bevin P. Engelward, Andrew J. Dreslin, et al.   DNA Cross-Linking Agentsthe Toxicity and Clastogenicity of Certain Chemotherapeutic Mammalian 3-Methyladenine DNA Glycosylase Protects against

  Updated version

  http://cancerres.aacrjournals.org/content/58/17/3965

Access the most recent version of this article 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. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/58/17/3965To request permission to re-use all or part of this article, use this link

on March 16, 2020. © 1998 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from