[CANCER RESEARCH 58. 2070-2075, May 15. 1998]

Advances in Brief

Carcinogens Preferentially Bind at Methylated CpG in the p53 MutationalHot Spots1

James X. Chen, Yi Zheng, Melissa West, and Moon-shong Tang2

Dt'/iiirlnient of Carcinoxenexi*. University of Texas, M. D. Anderson Cancer Center. Science Park, Stnithville, Texas 78957

Abstract

The major mutational hot spots in human cancers occur at CpGsequences in the p53 gene. It is generally presumed that the majority ofmutations at these sites result from the endogenous deamination of methylated cytosine. Using a UvrABC incision method, we have found thatcytosine methylation greatly enhances guanine alkylation at all CpG sitesin l lie/o.l gene by a variety of carcinogens, including ben7.o(a)pyrene dialepoxide, benzo(£)chrysene diol epoxide, aflatoxin B, 8,9-epoxide, andiV-acetoxy-2-acetylaminofluorene. These findings suggest that mutational

hot spots at methylated CpG sequences in the p53 gene may be a consequence of preferential carcinogen binding at these sites.

Introduction

Mutations in proto-oncogenes and tumor suppressor genes that are

either hereditary or nonhereditary in origin are commonly observed inhuman cancers. Although proto-oncogenes can be activated by a

variety of genetic alterations, including point mutations, the positionand type of mutations that activate the oncogenic function of aproto-oncogene during carcinogenesis are generally dictated by their

conveyance of selective advantage and, thus, tend to be confined to afew sequence positions with limited variations (1). A much broaderspectrum and variety of signature mutations has been observed intumor suppressor genes (1-3), which is not surprising, considering

that there are more ways to abolish the function of a tumor suppressorgene product than there are ways to gain an oncogenic function froma proto-oncogene product. The frequency of mutation at each site

within a gene depends upon a number of factors, including the typeand frequency of DNA damage formation at that site, the rate andefficiency of repair of each type of damage, and the efficiency andfidelity of translesion synthesis by DNA polymerases; each of thesefactors may be affected by sequence context, and selective pressurefor phenotype may result in disproportionate representation of certainmutations.

It has been found that over 50% of human cancers have a mutationin the p53 tumor suppressor gene (2). Although these mutations aredistributed along the coding region of this gene at more than 200positions, most are located in the sequences which code for aminoacids in the DNA binding domain of the p53 protein (2, 3). Intrigu-

ingly. more than 30% of these p53 mutations occur at CpG sites incodons 157. 175, 245, 248, 273, and 282 (Fig. 1A), with up to 55% ofthe mutations in human colon cancers occurring at such sites (Fig. Iß;Refs. 2 and 3). Because all of the cytosine residues of CpG sites in thecoding region of the p53 gene are known to be methylated in a varietyof tissues (4). it has been hypothesized and is generally presumed thatmost of these mutations arise by endogenous cytosine deamination at

Received 3/27/98; accepted 3/30/98.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.

1This study was supported by Grants ES 03124 and ES 08389 from the National

Institutes of Environmental Health Sciences.2 To whom requests for reprints should be addressed.

the methylated CpG sites (5, 6). Consistent with this hypothesis is thefinding that a majority of the mutations observed in human cancers areC —¿�»T transitions (2, 3, 5-7). Recently, using the UvrABC nucleaseincision method in combination with ligation-mediated PCR tech

niques, we have found that, in lung cancer mutational hot spots atcodons 157, 248, and 273 of the p53 gene, are the preferential bindingsites for the activated cigarette smoke component BPDE3 (8). Our

findings raise the possibility that targeted carcinogen binding ratherthan selective advantage is the main determinant for mutational hotspots in the p53 gene in human cancers. Our further studies havedemonstrated that cytosine methylation at the CpG site determines thepattern of preferential BPDE binding at this site (9) and that BPDEadducts formed at mutational hot spots in the p53 gene are poorlyrepaired (10).

A significant portion of germ-line mutations in human hereditary

disorders and somatic mutations in human cancers, including most ofthe mutational hot spots in the p53 gene, occur at CpG dinucleotidesequences (2, 3, 5-7). In light of these findings, we have determined

the effect of cytosine methylation on the alkylation of guanine byvarious environmental and model carcinogens to better understand theetiology of human cancer. We have previously demonstrated that,under controlled conditions, UvrABC nuclease is able to incise bulkychemicals such as BPDE, NAAAF, 7,12-dimethylbenz(a)anthracenediol epoxide, and mitomycin C-induced DNA adducts both specifi

cally and quantitatively, as indicated by the quantitative relationshipsbetween the number of adducts formed in a defined DNA fragment,the number of UvrABC incisions induced by these adducts, and thenearly identical kinetics of UvrABC incisions at different sequences(11). We have used these same UvrABC nuclease reaction conditionsto determine the effect of cytosine methylation at the CpG site on theguanine alkylation by the bulky chemical carcinogens BPDE, AFB1,NAAAF, and BgCDE. These compounds were chosen not only because of their carcinogenic potency (12, 13) but also because theyinteract with different moieties within the guanine base structure(12-15) and, therefore, allow us to probe the nature of the effects of

methylation on guanine alkylation.

Materials and Methods

DNA Fragment Isolation and 32P Labeling. The ¡i53gene fragments

containing exons 5. 7. and 8 were isolated from />5.?-containing plasmids

(pAT153P53p, obtained from L. Crawford and S. P. Tuck, Imperial CancerResearch Fund Laboratories, London. UK) grown in Escherichia coli cells. Tolabel DNA fragments containing exon 8 sequence, the plasmid DNAs werelinearized by Avail digestion and then '"P-labeled at the 5' ends. The labeled

DNA fragments were then digested with a second restriction enzyme. S.v/)l. toproduce single end-labeled fragments. To label DNA fragments containing

exon 7 sequence, the plasmid DNAs were linearized by Avril and then<2P-labeled at the 5' ends. The labeled DNA fragments were then digested with

a second restriction enzyme. Apa\, to produce single end-labeled fragments.

1The abbreviations used are: BPDE. ben/.o(u)pyrene diol epoxide; NAAAF. /V-ace-

toxy-2-acetylaminofluorene: AFB1. aflatoxin B, 8.9-epoxide: BgCDE. benzo(/?)chrysenediol epoxide; SAM. S-adenosylmethionine.

2070

Research. on September 8, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

PREFERENTIAL CARCINOGEN BINDING AT CpG SEQUENCES

(0o«+^i>*—

o0)A3

Z350

-300

-250

-200

-150

-100

-50

-iA-I

IIIO.JIwJMJIlpIllJf^1^11gco •¿�* uì to h-eCOI^^JI^ifplrfll-Jiyf-^-^jIplI88 2 S §?•»-CM CM CM CM CM1

,nln.^l^fli^fOlSoo(O Is»

CM CM CMilSW1

,.PÕJfejJÉh.Lii.i«I. -fa.*» .«,-j .^-ÕCn

O »— CM O5CMCO CO CO CO CO

Codon Number

80-,^

70-O+5

60-3

50-H-

40•¿�0fl>

30"C

20-3Z

10-n

.BA.JJ^.Jl^ft.U, ,,A,,iJU,IUIAj/Aw,I.IWJ 1I ili frihi, ,o o oOÎ O i- § S 8

Codon NumberFig. 1. Frequency of p53 mulaiions in total human cancer (/t) and colon cancer (B) by codon position. Numbers were obtained from the p53 database (3).

at CpG sequences.. mutations occurring

5-C Cytosine Methylation at CpG Sites. The "P single end-labeled DNA

fragments were subjected to Sssi methylase treatment with SAM to methylateall cytosines at CpG sites according to vendor's instructions.

Carcinogen Modification of Methylated and Unmethylated DNA. DNAwith and without methylation treatment was then modified with BPDE (10~4mg/ml), BgCDE (5 x IO"5 mg/ml), AFB1 (IO"4 mg/ml), and NAAAF(2.5 x IO"1 mg/ml). The unreacted carcinogens were removed by diethyl ether

extractions and by ethanol precipitation of DNA.UvrABC Incision Assay. The adduci distributions at various sequence

positions were mapped by the UvrABC nuclease incision method. In brief,UvrABC nucleuses were added to carcinogen-modified DNA fragments at a

molar ratio of 6 in reaction buffer containing 100 mM KG. 1 mM ATP. 10 minMgCI2, 10 mM Tris (pH 7.5), and 1 mM EDTA. The reactions were carried outat 37°Cfor 60 min and were stopped by phenol and ether extractions followed

by ethanol precipitation. The resultant DNAs were denatured by dissolving in85% formamide and heating at 90°C for 4 min and were separated by

electrophoresis in 8% denaturing polyacrylamide gels.

Results and Discussion

DNA fragments containing p53 exon sequences were isolated fromplasmids, single 5' end-labeled with Õ2P,and then subjected to Sssl

methylase treatment with SAM to methylate all cytosines at CpG sites

(9). The DNA fragments with and without methylation treatment werethen modified with BPDE. NAAAF, AFB1. and BgCDE at differentconcentrations, and the adduct distributions at various sequence positions were mapped by the UvrABC nuclease incision method (11).The extent of cytosine methylation was determined by Maxam-Gilbertpyrimidine reaction. Because 5-C-methylated cytosines are not modified by hydrazine. both the 5' and 3' phosphodiester bonds of each

methylated cytosine are refractory to piperidine hydrolysis; therefore,no cytosine ladders should be observed at methylated cytosines (16).Fig. 2A (Lane 3) shows that, under our methylation conditions, all ofthe CpG sites in the DNA fragment containing exon 8 sequence of thep53 gene are methylated. As can be seen in Fig. 2A, the guanineresidues at all methylated CpG sites (codons 273, 282, 283, and 290)show great enhancement of alkylation by BPDE (Lane 8 versus Lane7). which binds guanine at the exocyclic amine; by AFB1 (Lane 12

versus Lane 11), which binds at the N7 position; and by NAAAF(Lane 14 versus Lane 13), which binds at the C8 position (12-15). Not

all CpG dinucleotides show the same enhancement for guanine alkylation; in fact, 2-5-fold variation was observed (Fig. 3). It appears that

the surrounding sequences of the CpG dinucleotide may play some

2071

Research. on September 8, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

PREFERENTIAL CARCINOGEN BINDING AT CpG SEQUENCES

Control y T A G BPDE B9CDE

+ + + + UvrABC- + - + SAM

AFB1 AAAF

III•¿�c-

l

*C-

•¿�c-

S3 "*•¿�

M . •¿�•¿�•--

290

- 3 | 283

G 282C

Fig. 2. The effect of cytosine methylaiion at the CpGdinuclcoiide tin the binding of bulky chemical carcinogens in the human p53 gene. The p53 gene fragmentscontaining exon 8 (A) and exon 7 (B) isolated fromp53-containing plasmids were 5'-single Õ2Pend-labeled.as described in "Materials and Methods." To methylate

the cytosines at the CpG siles. DNA fragments weretreated with S.v.vland SAM according to the vendor's

specifications (9). DNA with and without methylationtreatment was then modified with BPDH (H) 4 mg/ml),BgCDE (5 X 10'* mg/ml), AFB1 (10 4 mg/ml), andNAAAF (2.5 x 10~3 mg/ml) according to methods

described previously (27-29). These modified DNAfragments were then reacted with UvrABC nuclease, andIhe resultant DNAs were separated by electrophoresis inan 8% denatured polyacrylamide gel. as described previously (II. 27. 28). The gel was dried and exposed lo aphosphor screen. Because UvrABC nuclease incises 7nucleotides .V and 4 nucleotides _Vto a modified base, itsincision bands are 7 bases shorter than the correspondingMaxum and Gilbert purine bands. The hand intensityrepresents the extent of chemical modification at a particular sequence site. The sequences of the bands ofinterest are indicated. *C. methylated cytosine; Control,

DNA without chemical carcinogen modification.

*

I] G 273C

12345678 9 10 11 12 13 14

B+ + + -f + + UvrABC

- + SAMP P

Control T T A G BPDE BgCDE AFB1 AAAF

—¿� ZUG 248

•¿� —¿� —¿� 245

12345678 9 10 11 12 13 14

2072

Research. on September 8, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

PREFERENTIAL CARCINOGEN BINDING AT CpG SEQUENCES

B273

1.o.e0.6-....

1 ,. 1ia!• II1GAGGTGCGTGTTTGTGCCTGTCCTGGGAGAGA...I"1!

ilfri0.40.80.8-1

.1

•¿�0.8-0.6-0.4.

0.2.0

•¿�273

b,

.1.—. 1 , ill.290282

283ÕÕ

1CCGGCGCACAGAGGAAGAGAATCTCCGC"

1'•2901

'2822831

.1 ..GAGGTGCGTGTTTGTGCCTGTCCTGGGAGAGACCGGCGCACAGAGGAAGAGAATCTCCGC020.4.0.6.0.8-1

.rn

11 - i ¡p97^1ITII1?on

1

0.8'

0.6

0.4.

0.2-

282283

GAGGTGCGTGTTTGTGCCTGTCCTGGGAGAGACCGGCGCACAGAGGAAGAGAATCTCCGC

0.2

0.4

0.6-

0.8

1

FT1 'i I IPP1 I"1!

273

JU1

290

GAGGTGCGTGTTTGTGCCTGTCCTGGGAGAGACCGGCGCACAGAGGAAGAGAATCTCCGC

0.2-

0.4.

o.e-

o.tJ

•¿�T

2481 T

0.8-

0.6-

0.4-

0.2-

0\-

0

0.2

0.4

245

iLATGGGCGGCATGAACCGGAGGil-¡i| ^|p|¥

2481

0.8

0.6

0.4

0.2.

245

li IIIATGGGCGGCATGAACCGGAGG

0.20.60.81

.10.80.60.40.20.2-0.4.0.6-0.8-

1 .I"

I24r

r~245

Cih,ATGGGCGGCATGAACCÕ.

|| W ï1*> ]»I8nII.5GAGG|lI"

i

O 8

0.6

0.4

0.2

245 248

o

0.2

0.4

o.e

08

11

ATGGGCGGCATGAACCGGAGGIr r•¿�•

Fig. 3. Quaniitation of the effect of cytosine methylation ui CpG sites on the binding of polycyclic aromatic hydrcx'arhons in exon 8 (Ai and exon 7 (/f) of the human /»5.Õgene.

The intensity of carcinogen induced UvrABC incision bands was quantified by a Phosphorlmager (Molecular Dynamics) and normali/.ed by the amount of DNA applied in the gel.Yaxis, relative intensity at different sequences,which are indicated on the X axis. Top and bottom, methylated and unmclhyluted DNA. respectively, modified with BPDE («),AFB1(h), BgCDE (c), and NAAAF (d).

2073

Research. on September 8, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

PREFERENTIAL CARCINOGEN BINDING AT CpG SEQUENCES

role in this enhanced guanine alkylation. BgCDE, which binds at theexocyclic amine of both guanine and adenine (12-13), shows preferential binding only at the guanine residue at methylated CpG sites(Fig. 2A, Lane 10 versus Lane 9).

This cytosine methylation-enhanced carcinogen binding at the CpGsites was also observed in DNA fragments containing exon 5 (data notshown) and exon 7 (Fig. 2B) sequence. Results from Fig. 2B show thatBPDE, BgCDE, AFB1, and NAAAF bind 2-5-fold higher at methylated as compared to unmethylated CpG sites at codons 245 and 248of the p53 gene (Lane 8 versus Lane 7; Lane 10 versus Lane 9; Lane12 versus Lane 11', and Lane 14 versus Lane 13). It is worth notingthat, previously, using UvrABC incision in combination with ligation-mediated PCR, we were unable to detect BPDE-guanine adduci formation at the CpG site in codon 245, although the cytosine at this siteis methylated (8). Results in Fig. 2B, in contrast, clearly demonstratethat cytosine methylation greatly enhances BPDE-guanine binding atthe CpG site in this codon (Lane 8 versus Lane 7). This discrepancyis likely due to that the unknown structure at the surrounding sequences of codon 245 of the p53 gene allows UvrABC to incise BPDEadduct formed at this codon but does not allow efficient ligation withuniversal linker to occur, therefore resulting in poor amplification ofthe UvrABC incision-generated DNA fragment by the ligation-medi-ated PCR method. Cytosine methylation at CpG sites also has aneffect on the alkylation of guanines distant from CpG sequences. Mostnotably, it enhances guanine alkylation at codons 272 and 276 byNAAAF, but it reduces guanine alkylation by AFB1 at codons 276and 285 and by BPDE at codons 277 and 279.

Results presented in Figs. 2 and 3 suggest that cytosine methylationat CpG sites may dramatically change the stereo-structure and/orchemical environment of the guanines at such dinucleotide sequences;these changes may promote or enhance guanine interactions withbulky chemicals at CpG sites and either increase or decrease alkylation of guanines at sequences distant from CpG sites. It has beensuggested that cytosine methylation may increase the nucleophilicityof the exocyclic amine of its paired guanine and enhance its bindingwith electrophilic compounds (17); however, this interpretation doesnot provide a satisfactory explanation for the enhancement of alkylation at the N7 and C8 positions by AFB1 and NAAAF. Methylationof cytosine stabilizes (dC •¿�dG) polymer helix structure, and suchmethylated (dC •¿�dG) polymers have a high affinity for noncovalent

binding of the nonelectrophilic compound benzo(a)pyrene tetraol(18). However, it is not known whether a single cytosine-methylatedCpG dinucleotide has an increased affinity toward various electro-philes and whether enhanced noncovalent binding will lead to alkylation. Further elucidation of the cytosine methylation-induced structural changes at native sequences is needed to understand their effectson guanine alkylation.

It has been hypothesized that endogenous deamination of methylated cytosine at the CpG site may be the cause of the frequentmutations observed in human cancers (5, 6), particularly in thecase of colon cancers, in which 55% of the mutations in the p53gene occur at CpG sites (Table 1 and Fig. 1). The majority of themutations in the p53 gene of colon cancers are C —¿�»T transitionsat CpG sites, which are generally considered to be hallmarks ofmutations induced by deamination of methylated cytosine (Table1; Ref. 19). However, four important findings challenge the validity and the generality of this hypothesis. First, a significant proportion of mutations occurring at CpG sites of the p53 gene inlung, liver, head and neck, and breast involve G —¿�*T transversions(Table 1; Refs. 2 and 3); these mutations most likely are caused bymechanisms other than deamination of methylated cytosines. Second, it has been found that sequence context controls the conformation of the BPDE-DNA adduct (20, 21) and that different

Table 1 Types of base substitution mutations ai the CpG sequences of the p53 gene indifferent human cancers"

CancerLungLiverHead/neckBreastEsophagusColonHuman

cancersC^Tor

G->A63

(35%)23(48%)61

(63%)111(80%)54

(84%)285(94%)G-»T83

(46%)18(38%)23

(24%)17(12%)9(14%)15(5%)G-»C33(18%)7(15%)11(12%)11(8%)1(2%)4(1%)Total

mutationsalCpG35%18%30%30%40%55%37%

' These data were obtained from Ref. 3.

conformers of BPDE adducts at the same sequence may inducedifferent types of mutation (22, 23); for example, mild heat treatment of BPDE-adducted DNA enhances G —¿�»A transitions (C —¿�>

T in the complementary strand; Ref. 24), whereas polyethyleneglycol treatment enhances G —¿�»T transversions (25). It is conceivable that different kinds of adducts in different sequence contextsmay induce different kinds of mutations. Therefore, there is noa priori reason to conclude that C —¿�»T (G —¿�»A in the comple

mentary strand) transition at the CpG dinucleotide must be theresult of cytosine deamination. Third, we have shown that cytosinemethylation at CpG dinucleotides greatly enhances guanine alkylation by bulky chemical carcinogens at these sequences. Fourth,we have found that repair of bulky chemical induced DNA damageat these CpG sites in the nontranscribed strand of the p53 gene issignificantly slower than repair at other sites (10). Therefore, onthe basis of these findings, we propose that this higher affinity fora wide variety of DNA-damaging agents at methylated CpG sitesmay be the major reason that most mutations in human canceroccur at sequences containing CpG dinucleotides. It should benoted that this hypothesis does not preclude the possibility thatadducts formed at the CpG sites may have higher mutability thanthose formed at other sequences. The unique structure at a methylated CpG site may allow a variety of chemical modifications,with differential effects on the fidelity of translesion DNA synthesis. Thus, the type of mutations that occur at CpG sites may bedependent on the nature of the damaging agent and may vary fordifferent tissues and organs. DNA-damaging agents that cause lungcancer may induce predominantly G —¿�»T transversions, but thosethat cause colon cancer may induce mostly G —¿�>A transitions.Therefore, a signature of G —¿�»A transition mutations or G —¿�»Ttransversion mutations may not be a reliable indicator of whetherthe mutations are due to spontaneous deamination of methylatedcytosines or due to guanine modifications.

Polycyclic aromatic hydrocarbons and AFB1 contaminants arewidely found in food, air, and water (12-15). Our finding that muta-tional hot spots in the p53 gene in human cancers have a 2-5-foldenhanced affinity toward such carcinogens strongly suggests thatDNA damage induced by these agents, in addition to endogenousdeamination of methylated cytosines, may be responsible for most ofthe mutations leading to carcinogenesis. Deamination of methylatedcytosines at CpG sites has been implicated as the primary mechanismof gene dysfunction in colon cancer (6, 26). If spontaneous deamination of methylated cytosine is a major cause leading to mutations inhuman cancers, these mutations will be difficult to prevent. Ourfindings challenge these assumptions and provide further evidencethat reductions in carcinogen exposure may reduce cancer risk.

Acknowledgments

We thank Dr. Frederick Beland at the National Center for ToxicologicalResearch and Dr. Ronald Harvey at the University of Chicago for the generousgifts of NAAAF and BgCDE.

2074

Research. on September 8, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

PREFERENTIAL CARCINOGEN BINDING AT CpG SEQUENCES

References

1. Ruddon. R. W. Cancer Biology. New York: Oxford University Press. 1995.2. Greenblatt, M. S.. Bennett, W. P., Hollstcin. M., and Harris. C. C. Mutations in the

p53 tumor suppressor gene: clues to cancer etiology and molecular pathogenesis.Cancer Res., 54: 4855-4878. 1994.

3. Hollstein. M., Shomer, B.. Greenblatt, M., Soussi. T.. Hovig, E.. Montesano. R., andHarris. C. C. Somatic point mutations in the p53 gene of human tumors and cell lines:updated compilation. Nucleic Acids Res.. 24: 141-146, 1996.

4. Tornaletti. S., and Pfeifer, G. P. Complete and tissue-independent methylation of CpG

sites in p53 gene: implications for mutations in human cancers. Oncogene. 10:1493-1499. 1995.

5. Rideout, W. M., III. Coet7.ee. G. A., Olumi, A. F., and Jones, P. A. 5-Methylcytosine

as an endogenous mutagen in the human LDL receptor and p53 genes. Science(Washington DC). 249.- 1288-1290. 1990.

6. Jones. P. A.. Buckley. J. D.. Henderson, B. E.. Ross. R. K., and Pike. M. C. From geneto carcinogen: a rapidly evolving field in molecular epidemiology. Cancer Res.. 51:3617-3620. 1991.

7. Cooper, D. N., and Youssofian. M. The CpG dinucleotide and human genetic disease.Hum. Genet., 78: 151-155, 1988.

8. Denissenko, M. F., Pao. A., Tang. M-s.. and Pfeifer. G. P. Preferential formation ofbenzopyrene adducts as lung cancer mutational hotspots \np53. Science (WashingtonDC). 274: 430-432. 1996.

9. Denissenko. M. F., Chen, J. X., Tang. M-s.. and Pfeifer. G. P. Cytosine methylationdetermines hot spots of DNA damage in the human p53 gene. Proc. Nati. Acad. Sci.USA. 94: 3893-3898. 1997.

10. Denissenko. M. F., Pao. A.. Pfeifer. G. P.. and Tang, M-s. Slow repair of bulky DNA

adducts along the nontranscribed strand of the human p53 gene may explain the strandbias of transversion mutations in cancers. Oncogene. 16: 1241-1247. 1988.

11. Tang. M-s. Mapping and quantification of bulky-chemical-induced DNA damage

using UvrABC nucleases. In: G. Pfeifer (ed.). Technologies for Detection of DNADamage and Mutation, p. 139. New York Plenum Press. 1996.

12. Harvey, R. G. Polycyclic Aromatic Hydrocarbons: Chemistry and Carcinogenicity.Cambridge, United Kingdom: Cambridge University Press, 1991.

13. Groopman, J. D.. and Cain. L. G. Interactions of fungal and plant toxins with DNA:aflatoxins, sterigmacystin. safrole. cycasin. and pyrrolizidine alkaloids. In: C. S.Cooper and Graver. P. L. (eds.). Chemical Carcinogenesis and Mutagenesis. pp.373-407. Berlin: Springer-Verlag. 1990.

14. Friedberg. E. C., Walker. G. C., and Siede. W. DNA Repair and Mutagenesis.Washington. DC: ASM Press. 1995.

15. Essigmann. J. M.. Groy. R. G.. Nadzan, A. M.. Busby. W. F.. Reinhold, V. N..Buschi. G., and Wogan. G. N. Structural identification of the major DNA adduciformed by aflatoxin B, in vitro. Proc. Nati. Acad. Sci. USA, 74: 1870-1874.

1977.

16. Church. G. M.. and Gilben. W. Genomic sequencing. Proc. Nati. Acad. Sci. USA. 81:1991-1995. 1984.

17. Johnson, W. S.. He, Q-Y., and Tomas/, M. Selective recognition of the m5CpG

dinucleolide sequence in DNA by mitomycin C for alkylation and cross-linking.Bioorg. Med. Chem.. 3: 851-860. 1995.

18. Geacintov. N. E., Shahba/, M.. Ibane/.. V.. Moussaoui, K., and Harvey. R. G. Base-sequence dependence of noncovalent complex formation and reactivity of benzo|ii|pyrenediol epoxide with polynucleolides. Biochemistry. 27: 8380-8387, 1988.

19. Coulondre. C., Miller. J. M.. Farabaugh. P. J.. and Gilbert. W. Molecular basis of basesubstitution hotspots in Escherichia coli. Nature (Lond.). 274.- 775-780, 1978.

20. Marsch. G. A.. Jankowiak, R., Suh, M.. and Small, G. J. Sequence dependence ofbenzopyrene diol epoxide-DNA adduci conformer distribulion: a sludy by laser-

induced fluorescence/polyacrylamide gel eleclrophoresis. Chem. Res. Toxicol., 7:98-109. 1994.

21. Geacintov. N. E., Cosman. M.. Hingcrty. B. E.. Amin. S., Broyde. S.. and Palei. D.NMR solution structures of slereoisomeric covalenl polycyclic aromatic cacinogcn-DNA adducts: principles, pallems, and diversity. Chem. Res. Toxicol., 10: 111-146,

1997.22. Rodriguez, H.. and Loechler, E. L. Mutational specificity of the ( + )-<mri-diol epoxide

of benzopyrene in a xupF gene of an Escherichia culi plasmid: DNA sequence contextinfluences hotspots. mutagenic specificily and the extenl of SOS enhancement ofmutagenesis. Carcinogenesis (Lond.), 14: 373-383. 1993.

23. Rodriguez. H.. and Loechler, E. L. Mutagenesis by the ( + )-rtnl/-diol epoxide of benzopyrene: what controls mulagenic specificily? Biochemislry. 32: 1759-1769, 1993.

24. Jelinsky. S. A.. Liu. T., Geacinlov. N. E., and Loechler, E. L. The major N2-Guaadduci of the ( + )-rtnf/-benzo|ii)pyrenc diol epoxide is capable of inducing G —¿�»A andG -»C. in addition lo G —¿�T. mulalions. Biochemistry. 34: 13545-13553, 1995.

25. Jelinsky. S., Shukla. R., Kozack. R., Geacintov, N.. and Loechler. E. L. Benzo-|«|pyrene ( + )-<mii-7.8-diol-9,10-oxide mutagenic mechanisms. Proc. Am. Assoc.

Cancer Res., 38: 127. 1997.26. Schmutte. C.. Yang. A. S.. Nguyen. T. T., Beart, R. W.. and Jones, P. A. Mechanisms

for the involvemenl of DNA methylation in colon Carcinogenesis. Cancer Res., 56:2375-2381. 19%.

27. Tang, M-s.. Pierce. J. R.. Doisy. R. P., Nazimicc, M. E.. and MacLeod, M. C.

Differences and similarities in Ihe repair of Iwo henzo[i/]pyrene diol epoxide isomersinduced DNA adducls by uvrA. u\rH. and u\rC gene producís. Biochemislry, 31 :8429-8436. 1992.

28. Pierce, J. R.. Case. R.. and Tang. M-s. Recognition and repair of 2-aminofluorene-and 2-(acetylamino)tluorene-DNA adducls by UVRABC nuclcase. Biochemislry. 28:5821-5826. 1989.

29. Agarwal. R.. Coffing. S. L.. Baird. W. M.. Kiselyov. A. S.. Harvey. R. G.. and Dipple.A. Metabolic activation of benzoli;(chrysene in Ihe human mammary carcinoma cellline MCF-7. Cancer Res.. 57: 415-419. 1997.

2075

Research. on September 8, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from

1998;58:2070-2075. Cancer Res   James X. Chen, Yi Zheng, Melissa West, et al.   Mutational Hot SpotsCarcinogens Preferentially Bind at Methylated CpG in the p53

  Updated version

  http://cancerres.aacrjournals.org/content/58/10/2070

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/58/10/2070To request permission to re-use all or part of this article, use this link

Research. on September 8, 2020. © 1998 American Association for Cancercancerres.aacrjournals.org Downloaded from