Tobacco smoke carcinogens, DNA damage and p53 …

Tobacco smoke carcinogens, DNA damage and p53 mutations in

smoking-associated cancers

Gerd P Pfeifer*,1, Mikhail F Denissenko2, Magali Olivier3, Natalia Tretyakova4,Stephen S Hecht*,4 and Pierre Hainaut*,3

1Division of Biology, Beckman Research Institute of the City of Hope, Duarte, California, CA 91010, USA; 2Sequenom Inc., SanDiego, California, CA 92121, USA; 3International Agency for Research on Cancer (WHO), 150 Cours Albert Thomas, 69372Lyon cedex, France; 4University of Minnesota Cancer Center, Mayo Mail Code 806, 420 Delaware St. S.E., Minneapolis,Minnesota, MN 55455, USA

It is estimated that cigarette smoking kills over 1 000 000people each year by causing lung cancer as well as manyother neoplasmas. p53 mutations are frequent in tobacco-related cancers and the mutation load is often higher incancers from smokers than from nonsmokers. In lungcancers, the p53 mutational patterns are differentbetween smokers and nonsmokers with an excess of Gto T transversions in smoking-associated cancers. Theprevalence of G to T transversions is 30% in smokers’lung cancer but only 12% in lung cancers of nonsmokers.A similar trend exists, albeit less marked, in laryngealcancers and in head and neck cancers. This type ofmutation is infrequent in most other tumors aside fromhepatocellular carcinoma. At several p53 mutationalhotspots common to all cancers, such as codons 248 and273, a large fraction of the mutations are G to T eventsin lung cancers but are almost exclusively G to Atransitions in non-tobacco-related cancers. Two impor-tant classes of tobacco smoke carcinogens are thepolycyclic aromatic hydrocarbons (PAH) and thenicotine-derived nitrosamines. Recent studies have indi-cated that there is a strong coincidence of G to Ttransversion hotspots in lung cancers and sites ofpreferential formation of PAH adducts along the p53gene. Endogenously methylated CpG dinucleotides arethe preferred sites for G to T transversions, accountingfor more than 50% of such mutations in lung tumors.The same dinucleotide, when present within CpG-methylated mutational reporter genes, is the target ofG to T transversion hotspots in cells exposed to themodel PAH compound benzo[a]pyrene-7,8-diol-9,10-epoxide. As summarized here, a number of other tobaccosmoke carcinogens also can cause G to T transversionmutations. The available data suggest that p53 mutationsin lung cancers can be attributed to direct DNA damagefrom cigarette smoke carcinogens rather than toselection of pre-existing endogenous mutations.Oncogene (2002) 21, 7435 – 7451. doi:10.1038/sj.onc.1205803

Keywords: cigarette smoke; lung cancer; DNA damage;p53; carcinogens

Introduction

Cigarette smoking causes 30% of all cancer deaths indeveloped countries (World Health Organization,1997). In addition to lung cancer, cigarette smokingis an important cause of esophageal, oral, orophar-yngeal, hypopharyngeal, and laryngeal cancers as wellas pancreatic cancer, bladder cancer, and cancer of therenal pelvis (International Agency for Research onCancer, 1986b). Cigarette smoking has also beenlinked to cancers of the stomach, renal body, liver,colon, nose, and myeloid leukemia, although theconnection to these cancers is weaker (Chao et al.,2000; Doll, 1996). Carcinogenic compounds in cigar-ette smoke are thought to be responsible for thesecancers.

The mainstream smoke emerging from the mouth-piece of a cigarette is an aerosol containing about1010 particles/ml and 4800 compounds (Hoffmannand Hecht, 1990). Experimentally, vapor-phasecomponents of the smoke can be separated fromthe particulate phase by a glass fiber filter. The vaporphase comprises over 90% of the mainstream smokeweight (Hoffmann et al., 2001). The main constitu-ents of the vapor phase are nitrogen, oxygen, andcarbon dioxide. Potentially carcinogenic vapor phasecompounds include nitrogen oxides, isoprene, buta-diene, benzene, styrene, formaldehyde, acetaldehyde,acrolein, and furan (Hoffmann et al., 2001). Theparticulate phase contains at least 3500 compoundsand many carcinogens including polycyclic aromatichydrocarbons (PAH), N-nitrosamines, aromaticamines, and metals (Hoffmann et al., 2001). Cigarettesmoke condensate can be prepared by trapping non-volatile (mainly particulate phase constituents) incold-traps. Cigarette smoke condensate reproduciblyand robustly causes tumors when applied to mouseskin and implanted in rodent lung (InternationalAgency for Research on Cancer, 1986b). Fractions ofthe condensate which contain PAH also induce

*Correspondence: Gerd P Pfeifer; E-mail: [email protected]; StephenS Hecht; E-mail: [email protected]; Pierre Hainaut; E-mail:[email protected]

Oncogene (2002) 21, 7435 – 7451ª 2002 Nature Publishing Group All rights reserved 0950 – 9232/02 $25.00

www.nature.com/onc

tumors in these models, but the concentrations of thePAH are too low to explain the carcinogenicity(reviewed in Rubin, 2001). Other fractions of thecondensate have tumor promoting and cocarcinogenicactivity which enhance the carcinogenicity of thePAH-containing fractions. Inhalation experimentsusing Syrian Golden hamsters demonstrate that wholecigarette smoke and its particulate phase consistentlyinduce preneoplastic lesions and benign and malig-nant tumors of the larynx (International Agency forResearch on Cancer, 1986b). This model system hasbeen widely applied and is the most reliable one forinduction of tumors by inhalation of cigarette smoke.Tumors are also observed in hamsters exposed onlyto the particulate phase of smoke (InternationalAgency for Research on Cancer, 1986b). Recently,studies in A/J mice exposed to environmental tobaccosmoke (comprised of 89% mainstream smoke and11% sidestream smoke in this experimental model) byinhalation demonstrate a small but reproducibleincrease in lung tumor multiplicity (Witschi, 2000).Tumor induction in this model is due to vapor phaseconstituents of cigarette smoke. Thus, there is reliableevidence that both particulate phase and vapor phaseconstituents of cigarette smoke cause tumors inlaboratory animals, and that tumor promoters andco-carcinogens are also involved in the observedresponse.

There are over 60 carcinogens in cigarette smokethat have been evaluated by the International Agencyfor Research on Cancer, and for which there is‘sufficient evidence for carcinogenicity’ in eitherlaboratory animals or humans (Hoffmann et al.,2001). They belong to various classes of chemicals, asfollows: PAH (10 compounds), aza-arenes (3), N-nitrosamines (8), aromatic amines (4), heterocyclicamines (8), aldehydes (2), volatile hydrocarbons (4),nitro compounds (3), miscellaneous organiccompounds (12), and metals and other inorganiccompounds (9). Other carcinogens not evaluated byIARC are also likely to be present. For example,among the PAH, multiple alkylated and high mole-cular-weight compounds have been detected, but areincompletely characterized with respect to theircarcinogenicity (Snook et al., 1978). Eighteen N-nitrosamines are present in cigarette smoke (Hoffmannet al., 2001). Published lists also include 106 aldehydesand 138 monocyclic aromatics; some may be carcino-genic (Hoffmann et al., 2001).

Tobacco smoke carcinogens and cancer

The potential role of tobacco smoke carcinogens insmoking-associated cancers can be evaluated byvarious means, but it is important to consider levelsof the compounds in cigarette smoke and their abilityto induce tumors in laboratory animals. In thefollowing, we discuss these factors with respect tocancers of the lung, oral cavity, esophagus, pancreas,and bladder.

Established pulmonary carcinogens in cigarettesmoke include PAH, aza-arenes, tobacco-specificnitrosamines, e.g. 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), 1,3-butadiene, ethyl carbamate,ethylene oxide, nickel, chromium, cadmium, polonium-210, arsenic, and hydrazine. These compounds convin-cingly induce lung tumors in at least one animal speciesand have been positively identified in cigarette smoke.

Among the PAH, benzo[a]pyrene (BaP) is the mostextensively studied compound (Phillips, 1983; BesaratiNia et al., 2002a) and its ability to induce lung tumorsupon local administration or inhalation has beenconvincingly established (Hecht, 1999; InternationalAgency for Research on Cancer, 1983). Lung tumorswere not observed when BaP was administered in thediet to B6C3F1 mice (Culp et al., 1998). In studies oflung tumor induction by implantation in rats, BaP ismore carcinogenic than the benzofluoranthenes orindeno[1,2,3-cd]pyrene (Deutsch-Wenzel et al., 1983).Extensive analytical data convincingly demonstrate thepresence of BaP in cigarette smoke. Its sales-weightedconcentration in current ‘full-flavored’ cigarettes isabout 9 ng per cigarette (Chepiga et al., 2000). Theabundant literature on BaP tends to diminish attentionto other PAH such as dibenz[a,h]anthracene, 5-methylchrysene, and dibenzo[a,i]pyrene which aresubstantially stronger lung tumorigens than BaP inmice or hamsters, but occur in lower concentrations incigarette smoke than does BaP (Nesnow et al., 1995;Sellakumar and Shubik, 1974).

Among the N-nitrosamines, N-nitrosodiethylamine isan effective pulmonary carcinogen in the hamster, butnot the rat (Reznik-Shuller, 1983; International Agencyfor Research on Cancer, 1978). Its levels in cigarettesmoke (up to 3 ng/cigarette) are low compared to thoseof other carcinogens. The tobacco-specific N-nitrosa-mine NNK is a potent lung carcinogen in rodents(Hecht, 1998). Its activity is particularly impressive inrats, where total doses as low as 6 mg/kg, administeredby s.c. injection, or 35 mg/kg administered in thedrinking water, produced significant incidences of lungtumors. Even lower doses induced lung tumors whenconsidered in dose-response trend analyses (Hecht,1998). It is the only compound in cigarette smokeknown to induce lung tumors systemically in all threecommonly used rodent models. NNK has a remarkableaffinity for the lung, causing mainly adenoma andadenocarcinoma, independently of the route of admin-istration (Hecht, 1998). NNK is the most abundantsystemic lung carcinogen in cigarette smoke. Multipleinternational studies definitively document the presenceof NNK in cigarette smoke; its sales-weighted concen-tration in current ‘full-flavored cigarettes’ is 131 ng/cigarette (Chepiga et al., 2000; Spiegelhalder andBartsch, 1996; Hecht and Hoffmann, 1988).

Lung is one of the multiple sites of tumorigenesis by1,3-butadiene in mice, but is not a target in the rat(International Agency for Research on Cancer, 1992).B6C3F1 mice develop lung tumors at exposureconcentrations that are three orders of magnitudelower than those that cause cancer in Sprague-Dawley

Genotoxicity of tobacco smokeGP Pfeifer et al

7436

Oncogene

rats. These interspecies differences are likely due todifferences in metabolism of 1,3-butadiene. Miceconvert a higher portion of the parent compound tohighly carcinogenic 1,2,3,4-diepoxybutane, while thedetoxification pathway via conjugation withglutathione is more prominent in rats (Thornton-Manning et al., 1995). Ethyl carbamate is a well-established pulmonary carcinogen in mice but not inother species (International Agency for Research onCancer, 1974a). Ethylene oxide induces pulmonarytumors in mice, but not in rats (International Agencyfor Research on Cancer, 1986a). Nickel, chromium,cadmium, and arsenic are all present in tobacco and apercentage of each is transferred to mainstream smoke(Hoffmann et al., 2001). Levels of polonium-210 intobacco smoke are insufficient to have a significantimpact on lung cancer initiation in smokers (Harley etal., 1980). Hydrazine is an effective lung carcinogen inmice and has been detected in cigarette smoke(International Agency for Research on Cancer, 1973).Formaldehyde and acetaldehyde induce nasal tumorsin rats when administered by inhalation (InternationalAgency for Research on Cancer, 1982, 1985, 1999;Swenberg et al., 1980). Although they are not lungcarcinogens, their concentrations in cigarette smoke areso high that they may nevertheless play a significantrole. There is approximately 100 000 times moreacetaldehyde in a cigarette than BaP (Chepiga et al.,2000).

Collectively, the available data indicate that PAHand NNK are important lung carcinogens in cigarettesmoke most likely to be involved in lung cancerinitiation in smokers. Their potent carcinogenicactivities compensate for their relatively low concentra-tions in tobacco smoke. Other carcinogens mentionedhere, as well as tumor promoters and co-carcinogens,may also play a role as causes of lung cancer insmokers.

The potent PAH carcinogen 7,12-dimethylbenz[a]an-thracene (DMBA) is routinely used for induction oforal tumors in the hamster (Solt et al., 1987). However,DMBA is not present in cigarette smoke. Other PAHhave been less frequently tested in this model. Amixture of NNK and NNN induced oral tumors in ratstreated repetitively by oral swabbing (Hecht et al.,1986). The rat oral cavity is one target of benzenecarcinogenecity (National Toxicology Program 1986).The risk for oral cancer is markedly enhanced byalcohol consumption in smokers, perhaps due in partto enhancement of carcinogen metabolic activation byethanol (Melikian et al., 1990; McCoy and Wynder,1979).

Numerous N-nitrosamines are potent esophagealcarcinogens in rats (Preussmann and Stewart, 1984).Among these, N’-nitrosonornicotine (NNN) is by farthe most prevalent in cigarette smoke. N-nitrosodiethy-lamine and N-nitrosopiperidine are two other smokeconstituents that could be involved in esophagealtumor induction in smokers. BaP induces someesophageal tumors when administered to mice in thediet (Culp et al., 1998). The risk for esophageal cancer

in humans is also enhanced by alcohol consumption(McCoy and Wynder, 1979).

NNK and its major metabolite 4-(methylnitrosami-no)-1-(3-pyridyl)-1-butanol (NNAL) are the onlyknown pancreatic carcinogens in cigarette smoke(Rivenson et al., 1988). Low doses of these nitrosa-mines induce pancreatic tumors in rats, in addition tolung tumors (Hecht, 1998). Pancreatic tumors are alsoobserved in the offspring of pregnant rats treated withNNK, and this effect is markedly enhanced by ethanol(Schuller et al., 1993).

4-Aminobiphenyl and 2-naphthylamine are knownhuman bladder carcinogens (International Agency forResearch on Cancer, 1972, 1974b). Both are present incigarette smoke. Hemoglobin adducts of 4-aminobi-phenyl and other aromatic amines are associated withbladder cancer induction in smokers (Castelao et al.,2001). The evidence is strong that aromatic aminesplay a significant role as causes of bladder cancer insmokers (Vineis et al., 2001).

Cigarette smoke contains free radicals and inducesoxidative damage (Arora et al., 2001; Pryor, 1997). Thegas phase of freshly generated cigarette smoke haslarge amounts of nitric oxide and other unstableoxidants (Hecht, 1999). The particulate phase ispostulated to contain long-lived radicals that mayundergo quinone-hydroquinone redox cycling (Pryor,1997). The presence of such free radicals and oxidantscan lead to oxidative DNA damage. However, the roleof oxidative damage in cancer induced by cigarettesmoke is unclear.

Formation and repair of DNA adducts

Carcinogens form the link between nicotine addictionand lung cancer (Figure 1) (Hecht, 1999). Nicotine isthe reason people continue to smoke in spite of thewell-known adverse health effects. Nicotine is not acarcinogen. However, the cigarette is a disastrousnicotine delivery device because the carcinogensdiscussed above accompany nicotine in each puff.Although the dose of each carcinogen per cigarette isquite small, the cumulative dose in a lifetime ofsmoking can be considerable.

The response of the organism to carcinogen exposureis similar to that for any other foreign compound ordrug. Cytochrome P450 enzymes catalyze addition ofan oxygen atom to the carcinogen, increasing its watersolubility and converting it to a form that is morereadily excretable (Guengerich, 2001). This ‘metabolicdetoxification’ process is further assisted by phase 2enzymes, which convert the oxygenated carcinogen to aform that is highly soluble in water (Armstrong, 1997;Burchell et al., 1997; Duffel, 1997). To the extent thatthis process is efficient, the organism will be protected.However, some of the intermediates formed by theinteraction of cytochrome P450 enzymes with carcino-gens are in fact quite reactive, generally possessing anelectrophilic (electron-deficient) center. Such intermedi-ates or metabolites can react with DNA, resulting in


7437

Oncogene

the formation of DNA adducts. This process whichconverts an unreactive carcinogen to a form that bindsto DNA is known as metabolic activation (Miller,1994). The balance between metabolic activation anddetoxification varies among individuals and is likely toaffect cancer risk because DNA adducts are central tothe carcinogenic process (Tang et al., 2001; Hecht,1999). Most cigarette smoke carcinogens requiremetabolic activation.

Elaborate DNA repair systems have evolved toeliminate DNA adducts from the genome (Hoeij-makers, 2001). For instance, the nucleotide excisionrepair pathway eliminates DNA adducts consistingpredominantly of base-attached larger chemical groups(so-called bulky DNA adducts), as well as intra- andinterstrand DNA crosslinks. Adducts of PAH arerepaired by nucleotide excision repair. The baseexcision repair systems are more geared towardsremoving defective DNA bases characterized byattachment of small chemical groups or basesfragmented by ionizing radiation or chemical oxida-tion. A specialized direct repair system acts through theenzyme O6-methylguanine DNA methyltransferase.This pathway is important for repair of the miscodingmethylated base O6-methylguanine. Effective repairprocesses should result in the rapid elimination andconsistent reduction of the cellular levels of DNAadducts. If unrepaired damage is still present duringDNA replication, it may either cause replicative DNApolymerases to stop at the site of a lesion (resulting inarrest of DNA replication, and cell death, orchromosomal aberrations). Alternatively, the poly-merases may bypass the altered base, with thepossibility of base misincorporation. In fact, thisproblem is so serious that all cells have evolved

specialized DNA polymerases that are able to bypassvarious types of DNA damage (Livneh, 2001). In somecases, these lesion bypass polymerases are endowedwith the property of being able to correctly bypassspecific types of lesions. An example is DNApolymerase eta (the product of the XPV gene), whichcorrectly bypasses thymine-thymine dimers (Johnson etal., 1999; Masutani et al., 1999). The importance ofthese polymerases in mutation avoidance is exemplifiedin the human genetic disorder xeroderma pigmentosumvariant (XPV) in which the XPV gene is defective,resulting in a greatly increased risk of developingsunlight-associated skin cancers. Human DNA poly-merase eta can bypass a template containing a (+)-trans-anti-benzo[a]pyrene-N2-dG adduct (BPDE-N2-dG), derived from BaP, and predominantly incorpo-rates an adenine. This specificity of nucleotideincorporation correlates well with the known mutationspectrum of BPDE-N2-dG lesions in mammalian cells(Zhang et al., 2000).

If DNA adducts are bypassed incorrectly by a DNApolymerase, mutations may arise (Figure 1). Theresulting alterations may lead to creation of a newphenotype and, if growth controlling genes areinvolved, to cellular transformation and the develop-ment of tumors. Protooncogenes and tumor suppressorgenes could be critical targets for carcinogens (Pfeiferand Denissenko, 1998; Hussain and Harris, 1998). Insmokers, there is a chronic barrage of metabolicallyactivated carcinogens which cause these multiplechanges (Figure 1). This constant assault on genes iscompletely consistent with genetic derangements lead-ing to six proposed hallmarks of cancer: self-sufficiencyin growth signals, insensitivity to anti-growth signals,evasion of apoptosis, tissue invasion and metastasis,

Figure 1 Scheme linking cigarette smoke carcinogens with multiple genetic changes in lung cancer. A key aspect is the chronicexposure of DNA to multiple metabolically activated carcinogens, leading to multiple adducts and their consequent mutations.The time periods and sequence of genetic events are uncertain


7438

Oncogene

sustained angiogenesis, and limitless replicative poten-tial (Hanahan and Weinberg, 2000).

Identification of a link between DNA damage andspecific mutations in tumor cells would strengthen theunderstanding to what extent elements of the environ-ment are responsible for tumorigenesis in humans. Thisidea is based on the knowledge that DNA adductsinduced by different mutagens may have significantlydifferent mutational properties. Adducts may form in abase and sequence-specific context. For example, aparticular adduct may induce predominantly G to Ctransversions within a particular sequence context, andif such mutations were found in one type of humantumor, such a carcinogen would become a suspect, inparticular if there is epidemiological evidence thatexposure to this agent may be involved in causing thistype of cancer.

DNA adducts: structure, detection, and mutationinduction

Table 1 summarizes information on structures of someDNA adducts that are formed from representativetobacco smoke carcinogens and related compounds.Available data on the detection of these adducts inlung DNA from smokers and some likely mutationsthat may result from their presence are also summar-ized.

Adduct structures

Bay region diol epoxides are among the principal PAHmetabolites involved in DNA adduct formation(Szeliga and Dipple, 1998; Conney, 1982). Each diolepoxide metabolite has four isomeric forms. There aretwo diastereomers in which the benzylic-OH of the diolis either on the opposite side (anti-) or the same side(syn-) of the molecule as the epoxide ring. Each ofthese diastereomers exists as a pair of enantiomers.Each of the four diol epoxides reacts with DNA todiffering extents, either at the exocyclic amino group ofdeoxyguanosine (N2-) or deoxyadenosine (N6-) (Szeligaand Dipple, 1998). Each reaction of the exocyclicamino group results in either trans- or cis-ring openingof the epoxide ring. Therefore, there are 16 possibleadducts of this type from each PAH diol epoxidemetabolite (Szeliga and Dipple, 1998). All adducts havebeen thoroughly characerized for multiple PAHmolecules. Table 1 illustrates only one adduct fromeach of two PAH in tobacco smoke: BaP and 5-methylchrysene (5-MeC). This is quantitatively themajor one in each case, but it should be noted that15 other adducts are formed from BaP-7,8-diol-9,10-epoxide and from 5-MeC-1,2-diol-3,4-epoxide. Inaddition, another set of 16 adducts can be formedfrom 5-MeC-7,8-diol-9,10-epoxide (Melikian et al.,1988). The diol epoxide pathway is not the onlymechanism for adduct formation from PAH. Depur-inating adducts have been detected as a result of oneelectron oxidation, and adducts resulting from quinone

formation have also been characterized (Casale et al.,2001; Penning et al., 1999). Adducts are also formedvia 9-hydroxy-BaP-4,5-oxide (Ross and Nesnow, 1999).

N-nitrosamines are metabolized to intermediates thatalkylate various positions of the DNA bases (Preuss-mann and Stewart, 1984). The most thoroughlyinvestigated are 7-alkylguanines and O6-alkylguanines,shown for N-nitrosodimethylamine and NNK in Table1. Other products include N-1, N-3, and N2-deoxygua-nosines, N-1, N-7, N-3, and N6-deoxyadenosines, O2-,O4, and N-3 thymidines, O2-, N-3, and N-4 deoxycy-tidines, and phosphotriesters (Singer and Grunberger,1983). NNK and NNN are metabolized to intermedi-ates that pyridyloxobutylate deoxyguanosine (Hecht,1998). N-nitrosopyrrolidine, a cyclic N-nitrosamine,displays alkylation chemistry that is somewhat differentfrom that of acyclic nitrosamines because the alkylat-ing intermediate is tethered to an aldehyde. A complexmixture of deoxyguanosine adducts is produced,among which the N2-tetrahydrofuranyl structurepredominates (Wang et al., 2001b).

Ethylene oxide behaves like a typical alkylatingagent, reacting primarily at N-7 of deoxyguanosine, asshown in Table 1, but also at other positions (Zhao etal., 1999). 1,3-Butadiene is metabolized to 3,4-epoxy-1-butene, 3,4-epoxy-1,2-butanediol, and 1,2,3,4-diepoxy-butane (Zhao et al., 2000; Koivisto et al., 1999; Koc etal., 1999; Tretyakova et al., 1997). Most of the DNAadducts arise from the reactions of the diol epoxide atthe N-7 position of guanine, N-3 adenine, and N6-adenine (Table 1). Multiple stereoisomers are formed.Acetaldehyde reacts with the exocyclic amino group ofdeoxyguanosine to give a Schiff base as the majoradduct (Wang et al., 2000a). Several other adductsincluding a G-G crosslink, shown in Table 1, have alsobeen identified. Crotonaldehyde produces cyclic 1,N2-deoxyguanosine adducts by Michael addition andSchiff base adducts by reaction of the aldehyde groupwith the exocyclic amino group of deoxyguanosine(Wang et al., 2001a; Chung et al., 1999). Other adductsare formed, with multiple stereoisomers, by dimers of3-hydroxybutanal, produced by hydration of crotonal-dehyde (Wang et al., 2000b).

Aromatic amines such as 4-aminobiphenyl andheterocyclic aromatic amines react with DNA mainlyat C-8 of deoxyguanosine via their N-hydroxy-metabolites (Delclos and Kadlubar, 1997). Adductshave also been observed at N2- of deoxyguanosine, O6-of deoxyguanosine, and N6-of deoxyadenosine.

Vinyl chloride is metabolized to chloroethylene oxidewhich reacts with DNA giving 7-oxoethyldeoxyguano-sine as a major product along with ‘etheno’ adductssuch as 3,N2-ethenodeoxyguanosine and 1,N6-etheno-deoxyadenosine, as shown in Table 1 (Swenberg et al.,1999; Nair et al., 1999). Ethyl carbamate is metabolizedto vinyl carbamate which similarly reacts giving 1,N6-ethenodeoxyadenosine (Guengerich and Kim, 1991).

2-Nitropropane is metabolized to intermediates thataminate deoxyguanosine at the C-8 and N2-positions(Sodum and Fiala, 1998). Radical oxidants in cigarettesmoke are believed to give rise to 8-oxodeoxyguanosine


7439

Oncogene

Table 1 Representative DNA adducts of some cigarette smoke carcinogens and related compounds: occurence and likely consequent mutations


7440

Oncogene

while nitric oxide yields deoxyoxanosine (shown inTable 1) along with other products (Burney et al.,1999; Asami et al., 1997). Recent studies demonstratethat 8-oxodeoxyguanosine is further oxidized byperoxynitrite to a variety of products including thoseillustrated in Table 1 (Henderson et al., 2002).

Adduct detection

Table 1 summarizes information on the detection ofspecific adducts in lung DNA from smokers. The BaPadduct shown in Table 1, BPDE-N2-dG, has been thesubject of numerous studies. Convincing evidence,obtained by HPLC with fluorescence detection ofBaP tetraols released upon acid hydrolysis, clearlydemonstrates the presence of BPDE-N2-dG in somesamples of human pulmonary DNA (Rojas et al., 1998;Kriek et al., 1998). Methods such as 32P-postlabelingand immunoassay have reported the presence of ‘PAH-DNA adducts’ or ‘aromatic DNA adducts’ in humanlung, but these are mainly uncharacterized (Santella,1999; Kriek et al., 1998). It is not certain that they arein fact derived from PAH. Gupta et al. (1999) havereported that adducts detected by 32P-postlabeling inlung DNA of cigarette smoke-exposed rats wereendogenous adducts enhanced by cigarette smoke. Noother specific PAH adducts have been detected withcertainty in human lung.

7-Methylguanine and 7-hydroxyethylguanine havebeen detected in human lung DNA by 32P-postlabelling(Zhao et al., 1999). Levels of 7-methylguanine arehigher in smokers than in nonsmokers in some but notall studies (Hecht and Tricker, 1999). Evidence has alsobeen presented for O6-methyl- and O6-ethyldeoxygua-nosine in human lung DNA, but confirmation by othermethods is lacking (Wilson et al., 1989). GC-MSanalysis of released 4-hydroxy-1-(3-pyridyl)-1-butanone(HPB) establishes the presence of NNK or NNN-derived pyridyloxobutyl DNA adducts in human lung(Foiles et al., 1991). These adducts, which are producedmainly by reaction with deoxyguanosine, are higher inlung DNA from smokers than nonsmokers, asexpected based on the specificity of NNK and NNNto tobacco products.

GC-MS analysis of 4-aminobiphenyl released fromhuman lung DNA provides evidence in support of thepresence of the C-8 adduct, as also indicated by 32P-postlabeling and immunoassay (Culp et al., 1997; Linet al., 1994). Levels of this adduct were not related tosmoking. One study demonstrated the presence of 8-oxodeoxyguanosine in human lung DNA using HPLCwith electrochemical detection (Asami et al., 1997).Levels were higher in smokers than in nonsmokers.

Collectively, the available data provide convincingevidence for the presence of certain adducts listed inTable 1 in human lung DNA. It is very likely thatmany of the other adducts are also present, but theavailable methodology is not sensitive or specificenough to detect them, or has not been applied yet.

Likely mutations

Table 1 summarizes data obtained mainly, althoughnot exclusively, from site-specific mutagenesis studies.The major adduct of BaP illustrated in Table 1produces GC?TA mutations (Kozack et al., 2000;Seo et al., 2000). In site-specific mutagenesis studies,this event was sequence dependent. This adductinduced 495% G?T mutations in one sequencecontext (5’-TGC) and approximately 95% G?Amutations in another context (5’-AGA). This mayresults from conformational complexities (Kozack etal., 2000; Seo et al., 2000). In chromosomal genes,racemic benzo[a]pyrene-7,8-diol-9,10-epoxide (BPDE)produces predominantly G to T mutations (Eisenstadtet al., 1982; Mazur and Glickman, 1988; Chen et al.,1990; Wei et al., 1993; Ruggeri et al., 1993; Yoon et al.,2001). Some PAH diol epoxides such as those derivedfrom benzo[c]phenanthrene react extensively at deox-yadenosine in DNA and consequently producesignificant levels of A mutations (Szeliga and Dipple,1998; Bigger et al., 1992).

7-Alkyldeoxyguanosines such as those derived fromN-nitrosodimethylamine, NNK, ethylene oxide, 1,3-butadiene, and vinyl chloride readily depurinate givingrise to abasic sites. Replication past abasic sites resultspredominantly in GC?TA mutations (Kunkel, 1984).Therefore, these cigarette smoke constituents can be

Table 1 (continued)


7441

Oncogene

expected to produce GC?TA mutations in pulmonaryDNA. However, apurinic sites may be repairedrapidly via the base excision repair pathway. Pyridy-loxobutylation of DNA gives both GC?TA andGC?AT mutations, as demonstrated by analysis ofras mutations in lung DNA of mice treated with amodel pyridyloxobutylating compound, NNKOAc(Ronai et al., 1993).

Site specific mutagenesis experiments in humanembryonic kidney cells have shown that the mispairingcharacteristics of O6-pyridyloxobutyldeoxyguanosineare comparable to that of O6-methyldeoxyguanosine,with a high number of G?A transitions and smalleramounts of G?T transversions observed (Pauly et al.,2002). GC?TAmutations are also the predominant onesobserved in studies of mutagenesis by N2-deoxyguano-sine adducts of 1,3-butadiene (Carmical et al., 2000),adducts of 4-aminobiphenyl (Melchior et al., 1994), and8-oxodeoxyguanosine (Moriya, 1993). Other studies ofbutadiene mutagenesis demonstrate the occurrence ofGC?AT, AT?TA, and GC?TA mutations (Recio etal., 2000). Products of the further oxidation of 8-oxodeoxyguanosine by peroxynitrite are highly efficientin producing GC?TA mutations (Henderson et al.,2002). Collectively, the available data indicate that manyDNA adducts associated with cigarette smoke exposuremay produce GC?TA mutations.

The spectrum of p53 mutations in smoking associatedlung cancers – clues to etiology?

Mutations in the p53 gene are a common occurrence inhuman tumors and are found in approximately 40% ofhuman lung cancers. p53 mutations are generally morecommon in smokers than in nonsmokers (Greenblatt etal., 1994; Hernandez-Boussard and Hainaut, 1998).Initial studies demonstrated that p53 mutations in lungcancer are different from those in other cancers andthat an excess of G to T transversions is characteristicfor these tumors (Hollstein et al., 1991). G to Ttransversions have been described as a molecularsignature of tobacco smoke mutagens in smoking-associated lung cancers (Greenblatt et al., 1994;Hainaut and Hollstein, 2000). It was shown that thereis an increased frequency of G to T transversions inlung cancers from smokers compared to lung cancersfrom nonsmokers and compared to most other cancers(reviewed in Greenblatt et al., 1994; Husgafvel-Pursiainen and Kannio, 1996; Hernandez-Boussardand Hainaut, 1998; Bennett et al., 1999; Hainaut andPfeifer, 2001; Vahakangas et al., 2001). However, suchconclusions have been questioned by two recent reports(Rodin and Rodin, 2000; Paschke, 2000). These reportshave prompted us to re-analyse this important issueusing the currently available literature (Hainaut andPfeifer, 2001; Hainaut et al., 2001). These new analysesconfirmed and extended the existence of a specificmutation pattern in lung cancers of smokers, andindicate that the recent reports by Rodin and Rodin(2000) and Paschke (2000) are in error.

Figure 2 shows that the mutational spectra in lungcancers from smokers and nonsmokers are clearlydifferent. Twelve per cent of the p53 mutations innonsmokers are G to T transversions. This figureincludes the mutation events scored as G to T and as Cto A, since the latter correspond to G to T changesoccurring on the non-coding, transcribed strand ofgenomic DNA (by convention, the base changesinduced by a mutation are read on the coding, non-transcribed strand). The difference between 12% G toT mutations in nonsmokers and 30% G to T insmokers is statistically highly significant (P50.001; Chisquare test). It is also important to note that thefrequency of G to T transversions is much higher inlung cancers than it is in any other tumor type exceptfor liver cancers associated with geographic areaswhere contamination of food with aflatoxins has beendemonstrated. In most internal cancers, not stronglylinked to tobacco consumption, such as brain, color-ectal, and breast cancers, the frequency of G to Tmutations is between 8 and 10 per cent (Figure 2). Thisis quite similar to the percentage of G to T mutationsfound in nonsmokers. Nonsmokers have an increasedlevel of G to A transitions (47% as opposed tosmokers 29%), a difference that is also statisticallyhighly significant. In Figure 2, we include categories ofboth ‘designated smokers’ (where the smoking status isindicated in the literature) and ‘all lung cancer casesminus nonsmokers.’ This is based on the knowledgethat 90% or more of all lung cancers occur in smokers(Proctor, 2001). As expected, the proportion of G to Ttransversions in all lung cancers (minus nonsmokers) isremarkably similar to that observed in designatedsmokers (Figure 2).

In recent years there has been a rise in theproportion of lung cancers that are of adenocarcinomahistological type. This trend has been attributed tochanges in cigarette design which may result in deeperinhalation as smokers compensate for lower levels ofnicotine. In addition, levels of NNK in cigarette smokehave increased while those of BaP have decreased. Inlaboratory animals, NNK induces adenocarcinoma ofthe lung whereas BaP generally produces squamous celltumors (Hoffmann et al., 2001).

To address the issue of whether the differenthistological types of lung cancer show differences intheir p53 mutational spectra, we have analysed theIARC TP53 mutation database separately for thesetumors (Figure 3). The frequencies of G to Ttransversions in the p53 database were 31% inadenocarcinomas, 28% in squamous cell carcinomas,26% in small cell lung cancers, and 34% in large cellcarcinomas. Thus, these different lung cancers exhibitsimilar p53 mutation spectra.

The type of base changes seen along the entire p53coding sequence is very different in lung cancercompared to other cancers. p53 mutations do notoccur in a random fashion along the coding sequencebut are typically clustered at so-called mutationhotspots. All these hotspots are within the DNAbinding domain of the p53 protein spanning approxi-


7442

Oncogene

Figure 2 Spectra of p53 mutations in human lung cancers. Data were obtained from the January 2002 update of the IARC TP53mutation database (http://www.iarc.fr/p53/Index.html). Cell-lines and metastatic cancers were excluded, as well as cases of radon-,asbestos-, and mustard gas-associated p53 mutations (see Hainaut and Pfeifer, 2001, for an exact specification of the mutation data).The total number of mutations are indicated in brackets. Del/ins/complex, deletions, insertions, and complex mutations; NS, non-smokers; CRC, colorectal cancer

Figure 3 Spectra of p53 mutations in different histological types of lung cancer. ADC, adenocarcinoma; SCC, squamous cell car-cinoma; LCC, large cell carcinoma; Small CC, small cell carcinoma. The dataset ‘All lung minus NS’ from Figure 2 was used andsubdivided by histological type (total number of mutations indicated in brackets)


7443

Oncogene

mately 180 amino acids from codon 120 to 300. Figure4 shows the distribution of all mutations (upper panels)and, specifically that of G to T transversions (lowerpanels), along the p53 gene in lung cancer (Figure 4a)and in brain/breast/colorectal cancers (Figure 4b).Different hotspots of G to T mutations are observedin brain/breast/colon compared to lung. These hotspotcodons are of particular interest since they may allow amore specific assignment to a particular carcinogen ifthat site is preferentially damaged or mutated by thecompounds in question. However, hotspot codons mayexist solely as a consequence of preferential phenotypicselection in tumors. To address this issue, we have

compared the mutational events in different types ofcancers at a number of common hotspot codons.

Figure 5 shows that the major lung cancer hotspots158, 245, 248, and 273 are commonly G to Ttransversions in lung cancer but are generally othermutation types (almost exclusively G to A) in internaltumors not associated with smoking. This does notagree with the preferential selection theory put forwardby Rodin and Rodin (2000). In this theory, cigarettesmoke acts as a physiological stress to expand apopulation of cells harboring endogenous mutations.What also does not agree with the Rodin model is thattumor promoters in cigarette smoke, which should acton pre-existing endogenous mutations, do not bythemselves cause cancer but require initiators (Rubin2001).

In Figure 5, we present data for five lung cancer p53mutational hotspots, codons 157, 158, 245, 248, and273. The tumor types analysed are lung, breast, colon,and brain. These tumors all have substantial numbersof total mutations in the p53 database (between 750and 1600). Codons 157 and 158 are common mutationsites in lung cancers (mostly G to T) but are much lessfrequently mutated in other tumors (see Figure 4).These two codons may be considered as mutationalhotspots specific for lung cancers of smokers. As rightlypointed out by Rodin and Rodin (2000), the mutagenicspectrum of codon 157 is quite limited since G to Tmight be the most common substitution that wouldresult in a p53 protein with a mutant residue at codon157. This fact is indeed sufficient to explain why codon157 is more often mutated in lung than in other cancers,since the chances of formation of a G to T mutation atthat codon are much greater in lung cancers than incancers less related to tobacco exposure (see below).

The fact that codons 248 and 273 are often thetargets of G to T transversions in lung cancers is aperfect illustration of the respective roles of mutagen-esis and selection in shaping the mutation spectrum ofp53. These two codons are the most commonlymutated ones in the entire p53 mutation database,and this is true in almost every type of human tumor.Mutation at these residues probably has a drastic effecton p53 function, since the two residues form contactsbetween the p53 protein and its DNA target (Walker etal., 1999). Mutating these residues abrogates thecapacity of p53 to act as a transcription factor toactivate certain downstream genes such as p21 or BAX.Importantly, in lung cancers, the mutations at thesecodons differ dramatically from those in cancers notlinked to tobacco smoking (Figure 5). At codons 248and 273, 35 – 45% of the mutations are G to Ttransversions in lung cancer, but this type of mutationis virtually absent in the other tumors. It should benoted that in 1994, Soussi and colleagues have reportedthat there were no substantial differences between thein vitro functional properties of G to T and G to Amutants at codons 248 and 273 (Ory et al., 1994).

These observations provide direct evidence that p53mutations in lung cancer occur by a distinct mechan-ism, and cannot be explained simply by selection.

Figure 4 Distribution of mutations along the p53 gene in lungcancer (a) and in breast/brain/colorectal cancers (b). The numberof p53 mutations (y axis) is shown by codon position (x axis; ma-jor peaks are labeled). Upper panel, distribution of all point mu-tations in all lung cancer cases excluding nonsmokers andoccupationally exposed individuals. Lower panel, distribution ofG to T transversions on the coding strand only. The spectrumof G to T transversions in lung cancers of nonsmokers is notshown since there were only 16 data points (one G to T mutationeach occurred at codons 135, 148, 158, 176, 198, 204, 237, 242,245, 273, 275 and 337, and four mutations occurred at codon 249)


7444

Oncogene

The distribution of G to T transversions in lung canceris consistent with the adduct spectrum and precisemutational specificity of PAH compounds

As indicated in the previous section, a characteristicmutational fingerprint in smoking-associated lungcancers is the high frequency of G to T transversions.PAHs are one class of carcinogens in tobacco smokethat produce predominantly this type of mutation invarious experimental systems (Eisenstadt et al., 1982;Mazur and Glickman, 1988; Chen et al., 1990; Wei etal., 1993; Ruggeri et al., 1993; Yoon et al., 2001). Asshown in Table 1, there are a number of other types ofDNA adducts derived from agents in tobacco smokethat also can give rise to G to T transversions. Theseinclude apurinic sites resulting from depurination of 7-alkylguanines, pyridyloxobutylated DNA, aromaticamine-DNA adducts, and 8-oxodeoxyguanosine andrelated products of oxidative damage.

The lung cancer p53 spectrum is consistent withthe mutational patterns induced by certain PAHs.The distribution of benzo[a]pyrene-7,8-diol-9,10-epox-ide (BPDE) and other PAH diol epoxide adductswas mapped at nucleotide resolution along exons ofthe p53 gene in PAH-treated normal humanbronchial epithelial cells (Denissenko et al., 1996;Smith et al., 2000). Frequent adduct formationoccurred at guanine positions in codons 157, 158,245, 248, and 273. These data are schematicallysummarized in Figure 6. These same positions ofpreferential PAH adduct formation are major muta-

tional hotspots in human lung cancers from smokers(see Figure 4a).

The distribution of BPDE-N2-dG within p53 exonfive was analysed using stable isotope labeling liquidchromatography- electrospray ionization tandem massspectrometry (HPLC-ESI-MS/MS) (Tretyakova et al.,unpublished results). In this approach, specific guaninenucleobases within p53 gene sequences are labeled with15N, so that the BPDE adducts originating from thesepositions can be distinguished from the lesions formedat other sites. An excellent agreement with the dataobtained by the UvrABC incision method wasobserved. Although the number of adducts did notdirectly correlate with mutations (for example, thehighest number of adducts was produced at codon 156,which is not a hotspot for mutations in lung cancer),these data did demonstrate an increased formation ofBPDE adducts at codons 156, 157, and 158. Otherfactors, e.g. DNA sequence effects on repair ormispairing, and lack of selection at codon 156 (thereis only a single G to T transversion reported at codon156 in the entire p53 database of more than 16 000mutations), may explain the predominance of muta-tions at codons 157 and 158. Alternatively, thesemutations may originate from DNA modification byother tobacco carcinogens, in particular other PAH.

The mechanistic basis for the selective occurrence ofsuch adduct hotspots in p53 gene is the enhancementof adduct formation by 5-methylcytosine bases presentat CpG dinucleotide sequences (Denissenko et al.,1997; Chen et al., 1998; Weisenberger and Romano,

Figure 5 Specific mutation profiles at common hotspot codons in lung and breast/brain/colorectal cancers. The percentages of thetotal number of mutations at each codon are given. Total numbers are: at codon 248, 67 (lung), 407 (brain/breast/colorectal); atcodon 273, 66 (lung), 366 (brain/breast/colorectal), at codon 245, 56 (lung), 171 (brain/breast/colorectal); at codon 158, 42 (lung),42 (brain/breast/colorectal); at codon 157, 40 (lung), 31 (brain/breast/colorectal). Data were obtained from the January 2002 updateof the IARC TP53 mutation database (http://www.iarc.fr/p53/Index.html)


7445

Oncogene

1999; Das et al., 1999; Pfeifer, 2000). All CpGsequences in the p53 coding exons five through nineare completely methylated in all tissues examinedincluding the lung (Tornaletti and Pfeifer, 1995).Methylation at CpG sites may increase the binding ofplanar carcinogen compounds at the intercalation step(Geacintov, 1986), although the precise mechanism bywhich cytosine methylation at CpG sites enhancescarcinogen binding still needs to be determined. Ourpreliminary data indicate that the increased reactivityof guanines in methylated CpG runs is primarily due tothe 5-methyl group at the base-paired cytosine (Matteret al., unpublished results). It has been shown that thepreferential formation of BPDE adducts at methylatedCpG sites is reflected in strongly enhanced mutagenesisat CpG sequences after treatment of cells with BPDE.This was demonstrated with three different CpG-methylated mutational reporter genes including twochromosomal genes that contain methylated CpGsequences (Yoon et al., 2001). The striking sequencespecificity of BPDE for producing G to T transversionhotspots at methylated CpG sequences is similar to thedistribution of G to T transversion hotspots insmoking-associated lung tumors (see Figure 4). In thep53 gene of lung cancer, five major G?T mutationalsites (in codons 157, 158, 245, 248, and 273) consist ofmethylated CpGs (Yoon et al., 2001).

It has been argued that methylated CpG sites arepreferentially modified and mutated by a range ofdifferent carcinogens including aromatic amines andaflatoxins (Chen et al., 1998; Hecht, 1999), however, theexact range of such compounds targeting methylatedCpGs is not known. We have not seen preferential

mutagenesis at methylated CpGs by the aromatic amine4-aminobiphenyl (Besarati Nia et al., 2002b).

Two recent studies lend support to the PAH – p53 -lung cancer connection. First, using a very sensitiveassay to detect mutations in the absence of selection,Hussain et al. (2001) have shown that exposure ofbronchial epithelial cells to BPDE produces G to Ttransversions in the p53 gene at lung cancer hotspotcodons 157, 248 and 249. Moreover, non tumorouslung tissues from smokers with lung cancer carry ahigh p53 mutational load at these codons, even whenanother mutation is present in the tumor itself. Thisobservation provides strong support to the idea thattobacco carcinogens can directly induce G to Ttransversions in exposed bronchial cells.

Second, DeMarini et al. (2001) determined p53 andKRAS mutations in lung tumors from 24 nonsmokingChinese women whose tumors were associated withexposure to smoky coal in unventilated homes. Thesewomen have high levels of exposure to various PAHs.The tumors showed a high percentage of mutationsthat were G to T transversions at either KRAS (86%)or p53> (76%). The mutations clustered at the CpG-rich codons 153 through 158 of the p53 gene, and atcodons 249 and 273 and had 100% of the guanines ofthe G to T transversions on the nontranscribed strand.This mutation spectrum is consistent with an exposureto PAHs (Smith et al., 2000; see also Figure 6), whichare present in substantial concentrations in smoky coalemissions.

The remarkable site specificity of mutagenesis byPAH compounds strongly suggests that targetedadduct formation in addition to phenotypic selection

Figure 6 PAH-diol epoxide adducts in p53 at lung cancer mutation hotspots in bronchial epithelial cells. Data were quantitatedfrom Smith et al. (2000). The length of the bars indicates the relative adduct frequency for each adduct at different sequence posi-tions in individual p53 exons. The strongest binding site is given a value of 1. The black bars indicate lung cancer mutational hot-spots as shown in Figure 4


7446

Oncogene

is responsible for shaping the p53 mutational spectrumin lung tumors. Furthermore, it is important to notethat the vast majority (90%) of G to T transversions inlung cancers are targeted to guanines on the non-transcribed DNA strand (IARC p53 mutation data-base). This is not the case for other mutation types, asfor example G to A transitions at CpG sites occur atthe same frequency on both strands. This observationimplies that, in the case of G to T transversions, astrand-specific DNA repair process plays a role in thepreferential repair of DNA lesions occurring on thetranscribed strand. DNA repair experiments analysingBPDE adducts in the p53 gene have shown that thenontranscribed strand is indeed repaired more slowlythan the transcribed strand (Denissenko et al., 1998).These findings support the proposal that both theinitial DNA adduct levels and a strand bias in repaircontribute to the mutational spectrum of the humanp53 gene in lung cancer.

The spectrum of p53 mutations in other smokingassociated cancers

In addition to lung cancers, several common neoplasmsare strongly associated with tobacco use. This is thecase for squamous cell carcinomas of the oral cavity,larynx and esophagus, and for cancers of the bladder(both squamous cell carcinomas and transitional cellcarcinomas). It is important to note that these cancersoccur at variable incidences in different regions of theworld, and that not all of these cancers are a direct

consequence of tobacco use. We have carried out ananalysis of the p53 mutation spectra in tumors, wherethere is strong evidence that tobacco smoking is animportant factor of risk (Figure 7).

Of the four cancer types, cancer of the larynx showsthe strongest similarities with lung cancers, with a highprevalence of G to T transversions (27%), many ofthem occurring at PAH-target codons (157,245).Strikingly, this is not the case for cancers at otherorgan sites. In oral and esophageal cancers, the p53mutation load has been shown to be proportional tothe extent of tobacco consumption, with an almostfourfold increase in mutation prevalence in heavysmokers as compared to nonsmokers (Brennan et al.,1995). In both cancers, however, the prevalence of G toT transversions is only slightly higher than in cancersnot strongly related to tobacco smoke (breast, color-ectal and brain cancers) and this difference is ofborderline statistical significance. A recent study notesthe similarity in mutational spectrum induced byacetaldehyde in the HPRT gene of human Tlymphocytes to that of the p53 gene in esophagealcancers (Noori and Hou, 2001). Moreover, thesetransversions do not preferentially occur at PAH-targetcodons. The patterns of mutations in both oral andesophageal cancers are extremely heterogeneous, inagreement with epidemiological data showing thatmultiple factors may act in conjunction in thepathogenesis of these cancers (including in particulartobacco and alcohol). Given the heterogeneity of themutation patterns, the data available at present do notallow the unambiguous identification of molecular

Figure 7 Spectra of p53 mutations in human cancers of the oral cavity, larynx, esophagus and bladder. Data were obtained fromthe January 2002 update of the IARC TP53 mutation database (http://www.iarc.fr/p53/Index.html). Cell-lines and metastatic can-cers were excluded, as well as cases of radon-, asbestos-, and mustard gass-associated p53 mutations. The total number of mutationsare indicated in brackets. Del/ins/complex, deletions, insertions, and complex mutations


7447

Oncogene

signatures of tobacco carcinogens in the p53 muta-tional spectrum of oral and esophageal cancers.

In the case of bladder cancers, the mutation patternshows an unusually high prevalence of G to Atransitions at non CpG sites. These mutations are notdistributed at random, and bladder-specific mutationhotspots have been reported at codons 280 and 285.Both of these codons occur within the same primarysequence context (5’AGAG), raising the possibility thatthis sequence is a preferential binding site for anexogenous agent involved in bladder carcinogenesis. Ithas been postulated that these mutations may representa fingerprint of aromatic amine which are the mostpotent class of bladder carcinogens in tobacco smoke.Indeed, codo 285 is a preferential binding site for N-hydroxy-4-amino=biphenyl (Feng et al., 2002).However, aromatic amines produce mainly G to Tmutations Besarati Nia et al., 2002).

Conclusions

The effect of individual exogenous agents in tobaccocarcinogenesis is difficult to assess at the molecularlevel because there is chronic exposure to a complexmixture of carcinogens, tumor promoters, and co-carcinogens. In the case of lung cancers, however, thereis strong evidence for the involvement of tobaccosmoke compounds at all steps of this chain of events.First, tobacco smoke compounds are absorbed andmetabolized in smokers. Second, many of thesecompounds produce DNA adducts in smokers’ lungs.Third, there is evidence that experimental exposure tometabolites of PAHs, one group of tobacco smokecarcinogens, can induce the same type of adducts incultured, normal bronchial cells. Fourth, G to Ttransversion mutations, consistent with the observedtype of adduct damage are detectable in both lungcancers and in adjacent, non-involved lung tissues insmokers. The position of these mutations oftencoincides with that of adducts detected in vitro. Fifth,the overall prevalence of p53 mutations is lower in lungcancers of nonsmokers than in smokers, and this isparticularly true for G to T mutations. Sixth, even at

codons that are common hotspots in all types ofcancers, there is an excess of G to T transversions inlung cancers of smokers as compared to nonsmokers orto cancers not directly related to tobacco use. Seventh,methylated CpG dinucleotides are the preferred sitesfor G to T transversions, accounting for more than50% of such mutations in lung tumors. The samedinucleotide, when present within mutational reportergenes, is the target of G to T transversion hotspots incells exposed to benzo[a]pyrene-7,8-diol-9,10-epoxide.

The fact that an apparent molecular signature ofPAHs is found in lung cancers does not rule out animportant role for other tobacco components incarcinogenesis. First, in lung cancers, other tobaccocarcinogens may produce a similar molecular signature,or may be responsible for non-G?T mutations.Second, different tobacco smoke compounds may exertcarcinogenic, co-carcinogenic, or tumor promotingeffects in an organ- and tissue specific manner,depending on the rate of accumulation and metabolismat various sites in the body. It is, however, interestingto note that, in the upper respiratory tract, there is agradient in the prevalence of p53 G to T transversionsin cancers of smokers, from low in the oral cavity, tointermediate in the larynx and high in varioushistological types of lung cancers. This situation mayreflect the existence of an underlying, parallel gradientin the extent of exposure of cells of the respiratorytract to PAHs and other tobacco smoke carcinogens.The studies on PAH-induced DNA damage and p53mutations provide a compelling link between a groupof exogenous carcinogens and human cancer. In thefuture, similar approaches on other tumors and othermutagens are expected to reveal further clues for therole of environmental mutagens in human carcinogen-esis.

Acknowledgments

The work of the authors was supported by grants from theNational Institutes of Health (CA84469 to GP Pfeifer; CA-81301 and DA 13333 to SS Hecht). The IARC p53database is supported by a grant from the EuropeanCommunity (QLG-1999-00273).

References

Armstrong RN. (1997). Comprehensive Toxicology:Biotransformation. Vol 3. Guengerich FP (ed). New York:Elsevier Science, pp. 307 – 327.

Arora A, Willhite C and Liebler D. (2001). Carcinogenesis,22, 1173 – 1178.

Asami S, Manabe H, Miyake J, Tsurudome Y, Hirano T,Yamaguchi R, Itoh H and Kasai H. (1997). Carcinogen-esis, 18, 1763 – 1766.

Bennett WP, Hussain SP, Vahakangas KH, Khan MA,Sheilds PG and Harris CC. (1999). J. Pathol., 187, 8 – 18.

Besarati Nia A, Kleinjans JC and Van Schooten FJ. (2002a).Biomarkers, 7, 209 – 229.

Besarati Nia ABates SEPfeifer GP (2002b). Cancer Res., 62,(in press).

Bigger C, St John J, Yagi H, Jerina D and Dipple A. (1992).Proc. Natl. Acad. Sci. USA, 89, 368 – 372.

Brennan JA, Boyle JO, Koch WM, Goodman SN, HrubanRH, Eby YJ, Couch MJ, Forastiere AA and Sidransky D.(1995). New Engl. J. Med., 332, 712 – 717.

Burchell B, McGurk K, Brierley C and Clarke D. (1997).Comprehensive Toxicology: Biotransformation. Vol 3.Guengerich FP (ed). New York: Elsevier Science, pp.401 – 436.

Burney S, Caulfield J, Niles J, Wishnok J and TannenbaumS. (1999). Mutation Res., 424, 37 – 49.

Carmical J, Zhang M, Nechev L, Harris C, Harris T andLloyd R. (2000). Chem. Res. Toxicol., 13, 18 – 25.


7448

Oncogene

Casale G, Singhal M, Bhattacharya S, Ramanathan R,Roberts K, Barbacci D, Zhao J, Jankowiak R, Gross M,Cavalieri E, Small G, Rennard S, Mumford J and Shen M.(2001). Chem. Res. Toxicol., 14, 192 – 201.

Castelao J, Yuan J, Skipper P, Tannenbaum S, Gago-Dominguez M, Crowder J, Ross R and Yu M. (2001). J.Natl. Cancer Inst., 93, 538 – 545.

Chao A, Thun M, Jacobs E, Henley S, Rodriguez C andCalle E. (2000). J. Natl. Cancer Inst., 92, 1888 – 1896.

Chen JX, Zheng Y, West M and Tang M-S. (1998). CancerRes., 58, 2070 – 2075.

Chen R-H, Maher VM and McCormick JJ. (1990). Proc.Natl. Acad. Sci. USA, 87, 8680 – 8684.

Chepiga T, Morton M, Murphy P, Avalos J, Bombick B,Doolittle D, Borgerding M and Swauger J. (2000). FoodChem. Toxicol., 38, 949 – 962.

Chung F, Zhang L, Ocando J and Nath R. (1999). ExocyclicDNA Adducts in Mutagenesis and Carcinogenesis. Singer Band Bartsch H (eds). Lyon, France: International Agencyfor Research on Cancer, pp. 45 – 54.

Conney A. (1982). Cancer Res., 42, 4875 – 4917.Culp S, Gaylor D, Sheldon W, Goldstein L and Beland F.(1998). Carcinogenesis, 19, 117 – 124.

Culp S, Roberts D, Talaska G, Lang N, Fu P, Lay Jr J, TeitelC, Snawder J, Von Tungeln L and Kadlubar F. (1997).Mutation Res., 378, 97 – 112.

Das A, Tan KS, Gopalakrishnan S, Waring MJ and TomaszM. (1999). Chem. Biol., 6, 461 – 471.

Delclos K and Kadlubar F. (1997). Comprehensive Toxicol-ogy: Chemical Carcinogens and Anticarcinogens. 12th edn.Bowden GT and Fischer SM (eds). pp. 141 – 170.

DeMarini DM, Landi S, Tian D, Hanley NM, Li X, Hu F,Roop BC, Mass MJ, Keohavong P, Gao W, Olivier M,Hainaut P and Mumford JL. (2001). Cancer Res., 61,

6679 – 6681.Denissenko MF, Chen JX, Tang MS and Pfeifer GP. (1997).Proc. Natl. Acad. Sci. USA, 94, 3893 – 3898.

Denissenko MF, Pao A, Pfeifer GP and Tang M-S. (1998).Oncogene, 16, 1241 – 1247.

Denissenko MF, Pao A, Tang M-S and Pfeifer GP. (1996).Science, 274, 430 – 432.

Deutsch-Wenzel R, Brune H, Grimmer G, Dettbarn G andMisfield J. (1983). J. Natl. Cancer Inst., 71, 539 – 543.

Doll R. (1996). Br. Med. J., 52, 35 – 49.Duffel M. (1997). Comprehensive Toxicology: Biotransfor-mation. Vol 3. Guengerich FP (ed). New York: ElsevierScience, pp. 365 – 384.

Eisenstadt E, Warren AJ, Porter J, Atkins D and Miller JH.(1982). Proc. Natl. Acad. Sci. USA, 79, 1945 – 1949.

Feng Z, Hu W, Rom WN, Beland FA and Tang M-S. (2002).Biochemistry, 41, 6414 – 6421.

Foiles PG, Akerkar SA, Carmella SG, Kagan M, Stoner GD,Resau JH and Hecht SS. (1991). Chem. Res. Toxicol., 4,364 – 368.

Geacintov NE. (1986). Carcinogenesis, 7, 759 – 766.Greenblatt MS, Bennett WP, Hollstein M and Harris CC.(1994). Cancer Res., 54, 4855 – 4878.

Guengerich F. (2001). Chem. Res. Toxicol., 14, 611 – 650.Guengerich F and Kim D. (1991). Chem. Res. Toxicol., 4,413 – 421.

Gupta R, Arif J and Gairola C. (1999). Mutation Res., 424,195 – 205.

Hainaut P and Hollstein M. (2000). Adv. Cancer Res., 77,81 – 137.

Hainaut P, Olivier M and Pfeifer GP. (2001). Mutagenesis,16, 551 – 553.

Hainaut P and Pfeifer GP. (2001). Carcinogenesis, 22, 367 –374.

Hanahan D and Weinberg R. (2000). Cell, 100, 57 – 70.Harley N, Cohen B and Tso T. (1980). Banbury Report 3: ASafe Cigarette? Gori GB and Beck FG (eds). Cold SpringHarbor Laboratory, pp. 93 – 104.

Hecht S and Hoffmann D. (1988). Carcinogenesis, 9, 875 –884.

Hecht S, Rivenson A, Braley J, DiBello J, Adams J andHoffmann D. (1986). Cancer Res., 46, 4162 – 4166.

Hecht S and Tricker A. (1999). III. Analytical Determinationof Nicotine and Related Compounds and Their Metabolites.Gorrod JW and Jacob P (eds). Amsterdam: ElsevierScience, pp. 421 – 488.

Hecht SS. (1998). Chem. Res. Toxicol., 11, 559 – 603.Hecht SS. (1999). J. Natl. Cancer Inst., 91, 1194 – 1210.Henderson P, Delaney J, Gu F, Tannenbaum S andEssigmann J. (2002). Biochemistry, 41, 914 – 921.

Hernandez-Boussard TM and Hainaut P. (1998). Environ.Health Perspect., 106, 385 – 391.

Hoeijmakers JH. (2001). Nature, 411, 366 – 374.Hoffmann D and Hecht S. (1990).Handbook of ExperimentalPharmacology. 94/1 ed. Cooper CS and Grover PL (eds).Heidelberg: Springer-Verlag, pp. 63 – 102.

Hoffmann D, Hoffmann I and El-Bayoumy K. (2001). Chem.Res. Toxicol., 14, 767 – 790.

Hollstein M, Sidransky D, Vogelstein B and Harris CC.(1991). Science, 253, 49 – 53.

Husgafvel-Pursiainen K and Kannio A. (1996). Environ.Health Perspect., 104, (Suppl. 3): 553 – 556.

Hussain SP, Amstad P, Raja K, Sawyer M, Hofseth L,Shields PG, Hewer A, Phillips DH, Ryberg D, Haugen Aand Harris CC. (2001). Cancer Res., 61, 6350 – 6355.

Hussain SP and Harris CC. (1998). Cancer Res., 58, 4023 –4037.

International Agency for Research on Cancer. (1972). IARCMonographs on the Evaluation of the Carcinogenic Risk ofChemicals to Man. Vol 1. Lyon, France: IARC.

International Agency for Research on Cancer. (1973). IARCMonographs on the Carcinogenic Risk of Chemicals to Man.Vol 4. Lyon, France: IARC, pp. 127 – 136.

International Agency for Research on Cancer. (1974a).IARC Monographs on the Carcinogenic Risk of Chemicalsto Man. Vol 7. Lyon, France: IARC, pp. 111 – 140.

International Agency for Research on Cancer. (1974b).IARC Monographs on the Evaluation of the CarcinogenicRisk of Chemicals to Humans. Vol 4. Lyon, France: IARC,pp. 97 – 111.

International Agency for Research on Cancer. (1978). IARCMonographs on the Evaluation of the Carcinogenic Risk ofChemicals to Humans. Vol 17. Lyon, France: IARC, pp.83 – 124.





7449

Oncogene

International Agency for Research on Cancer. (1986a).IARC Monographs on the Evaluation of the CarcinogenicRisk of Chemicals to Humans. Vol 60. Lyon, France:IARC, pp. 73 – 159.

International Agency for Research on Cancer. (1986b).IARC Monographs on the Evaluation of the CarcinogenicRisk of Chemicals to Humans. Vol 38. Lyon, France:IARC, pp. 37 – 375.



Johnson RE, Prakash S and Prakash L. (1999). Science, 283,1001 – 1004.

Koc H, Tretyakova N, Walker V, Henderson R andSwenberg J. (1999). Chem. Rev. Toxicol., 12, 566 – 574.

Koivisto P, Kilpelainen I, Rasanen I, Adler I, Pacchierotti Fand Peltonen K. (1999). Carcinogenesis, 20, 1253 – 1259.

Kozack R, Seo K, Jelinsky S and Loechler E. (2000).Mutation Res., 450, 41 – 59.

Kriek E, Rojas M, Alexandrov K and Bartsch H. (1998).Mutation Res., 400, 215 – 231.

Kunkel T. (1984). Proc. Natl. Acad. Sci. USA, 81, 1494 –1498.

Lin D, Lay JO, Bryant MS, Malaveille C, Friesen M, BartschH, Lang NP and Kadlubar FF. (1994). Environ. HealthPerspect., 102, (Suppl. 6): 11 – 16.

Livneh Z. (2001). J. Biol. Chem., 276, 25639 – 25642.Loechler E, Green C and Essigmann J. (1984). Proc. Natl.Acad. Sci. USA, 81, 6271 – 6275.

Masutani C, Kusumoto R, Yamada A, Dohmae N, YokoiM, Yuasa M, Araki M, Iwai S, Takio K and Hanaoka F.(1999). Nature, 399, 700 – 704.

Mazur M and Glickman B. (1988). Somat. Cell Mol. Genet.,14, 393 – 400.

McCoy G and Wynder E. (1979). Cancer Res., 39, 2844 –2850.

Melchior Jr W, Marques M and Beland F. (1994).Carcinogenesis, 15, 889 – 899.

Melikian A, Amin S, Huie K, Hecht S and Harvey R. (1988).Cancer Res., 48, 1781 – 1787.

Melikian A, Fudem Goldin B, Prahalad A and Hecht S.(1990). Chem. Res. Toxicol., 3, 139 – 143.

Miller J. (1994). Drug Metabol. Dispos., 26, 1 – 36.Misra R, Page J, Smith G, Waalkes M and Dipple A. (1998).Chem. Res. Toxicol., 11, 211 – 216.

Moriya M. (1993). Proc. Natl. Acad. Sci. USA, 90, 1122 –1126.

Nair J, Barbin A, Velic I and Bartsch H. (1999). MutationRes., 424, 59 – 69.

National Toxicology Program. (1986). NTP TechnicalReport No. 289, Toxicology and Carcinogenesis Studies ofBenzene (CAS No. 71-43-2) in F344/N Rats and B6C3F1

Mice (Gavage Studies). National Toxicology Program.Nesnow S, Ross J, Stoner G and Mass M. (1995).Toxicology, 103, 403 – 413.

Noori P and Hou S. (2001). Carcinogenesis, 22, 1825 – 1830.Ory K, Legros Y, Auguin C and Soussi T. (1994). EMBO J.,13, 3496 – 3504.

Paschke T. (2000). Mutagenesis, 15, 457 – 458.Pauly GT, Peterson LA and Moschel RC. (2002). Chem. Res.Toxicol., 15, 165 – 169.

Penning T, Burczynski M, Hung C, McCoull K, Palackal Nand Tsuruda L. (1999). Chem. Res. Toxicol., 12, 1 – 18.

Pfeifer GP. (2000). Mutat. Res., 450, 155 – 166.Pfeifer GP and Denissenko MF. (1998). Environ. Mol.Mutagen., 31, 197 – 205.

Phillips DH. (1983). Nature, 303, 468 – 472.Preussmann R and Stewart B. (1984). Chemical Carcinogens.2nd ed. Vol 2. ACS Monograph 182. Searle CE (ed).Washington, DC: American Chemical Society, pp. 643 –828.

Proctor RN. (2001). Nature Reviews Cancer, 1, 82 – 86.Pryor W. (1997). Environ. Health Perspect., 105, 875 – 882.Recio L, Saranko C and Steen A. (2000). Res. Rep. HealthEff. Inst. 49 – 87.

Reznik-Shuller H. (1983). Comparative Respiratory Tract-Carcinogens. 2nd edn. Reznik-Shuller HM (ed). BocaRaton, FL: CRC Press, pp. 109 – 159.

Rivenson A, Hoffmann D, Prokopczyk B, Amin S and HechtS. (1988). Cancer Res., 48, 6912 – 6917.

Rodin SN and Rodin AS. (2000). Proc. Natl. Acad. Sci. USA,97, 12244 – 12249.

Rojas M, Alexandrov K, Cascarbi I, Brockmoller J,Likhachev A, Pozharisski K, Bouvier G, Auburtin G,Mayer L, Koop-Schneider A, Roots I and Bartsch H.(1998). Pharmacogenetics, 8, 109 – 118.

Ronai ZA, Gradia S, Peterson LA and Hecht SS. (1993).Carcinogenesis, 14, 2419 – 2422.

Ross J and Nesnow S. (1999). Mutation Res., 424, 155 – 166.Rubin H. (2001). Carcinogenesis, 22, 1903 – 1930.Ruggeri B, DiRado M, Zhang SY, Bauer B, Goodrow T andKlein-Szanto AJP. (1993). Proc. Natl. Acad. Sci. USA, 90,1013 – 1017.

Santella R. (1999). Cancer Epidemiol. Biomarkers Prev., 8,733 – 739.

Schuller H, Jorquera R, Reichert A and Castonguay A.(1993). Cancer Res., 53, 2498 – 2501.

Sellakumar A and Shubik P. (1974). J. Natl. Cancer Inst., 53,1713 – 1719.

Seo K, Jelinsky S and Loechler E. (2000).Mutation Res., 463,215 – 246.

Singer B and Grunberger D. (1983). Molecular Biology ofMutagens and Carcinogens. New York: Plenum Press, pp.45 – 94.

Smith LE, Denissenko MF, Bennett WP, Li H, Amin S, TangM-S and Pfeifer GP. (2000). J. Natl. Cancer Inst., 92, 803 –811.

Snook M, Severson R, Arrendale R, Higman H and ChortykO. (1978). Beitrage Tabakforsch., 9, 222 – 247.

Sodum R and Fiala E. (1998). Chem. Res. Toxicol., 11,

1453 – 1459.Solt D, Polverini P and Claderon L. (1987). J. Oral Path., 16,294 – 302.

Spiegelhalder B and Bartsch H. (1996). Eur. J. Cancer Prev.,5, 33 – 38.

Swenberg J, Bogdanffy M, Ham A, Holt S, Kim A,Morinello E, Ranasinghe A, Scheller N and Upton P.(1999). IARC Sci. Publ., 29 – 43.

Swenberg J, Kerns W, Mitchell R, Gralla J and Pavkov K.(1980). Cancer Res., 40, 3398 – 3402.

Szeliga J and Dipple A. (1998). Chem. Res. Toxicol., 11, 1 –11.

Tang D, Phillips D, Stampfer M, Mooney L, Hsu Y, Cho S,Tsai W, Ma J, Cole K, She M and Perera F. (2001). CancerRes., 61, 6708 – 6712.


7450

Oncogene

Thornton-Manning JR, Dahl AR, Bechtold WE, Griffith JrWC and Henderson RF. (1995). Carcinogenesis, 16, 1723 –1731.

Tornaletti S and Pfeifer GP. (1995). Oncogene, 10, 1493 –1499.

Tretyakova N, Lin Y, Sangaiah R, Upton P and Swenberg J.(1997). Carcinogenesis, 18, 137 – 147.

Vahakangas KH, Bennett WP, Castren K, Welsh JA, KhanMA, Blomeke B, Alavanja MC and Harris CC. (2001).Cancer Res, 61, 4350 – 4356.

Vineis P, Marinelli D, Autrup H, Brockmoller J, Cascorbi I,Daly A, Golka K, Okkels H, Risch A, Rothman N, Sim Eand Taioli E. (2001). Cancer Epidemiol. Biomarkers Prev.,10, 1249 – 1252.

Walker DR, Bond JP, Tarone RE, Harris CC, MakalowskiW, Boguski MS and Greenblatt MS. (1999). Oncogene, 18,211 – 218.

WangM,McIntee E, Cheng G, Shi X, Villalta P and Hecht S.(2000a). Chem. Res. Toxicol., 13, 1149 – 1157.

WangM,McIntee E, Cheng G, Shi Y, Villalta P and Hecht S.(2000b). Chem. Res. Toxicol., 13, 1065 – 1074.

WangM,McIntee E, Cheng G, Shi Y, Villalta P and Hecht S.(2001a). Chem. Res. Toxicol., 14, 423 – 430.

Wang M, McIntee E, Shi Y, Cheng G, Upadhyaya P, VillaltaP and Hecht S. (2001b). Chem. Res. Toxicol., 14, 1435 –1445.

Wei SJ, Chang RL, Bhachech N, Cui XX, Merkler KA,Wong CQ, Hennig E, Yagi H, Jerina DM and Conney AH.(1993). Cancer Res., 53, 3294 – 3301.

Weisenberger DJ and Romano LJ. (1999). J. Biol. Chem.,274, 23948 – 23955.

Wilson VL, Weston A, Manchester DK, Trivers GE, RobertsDW, Kadlubar FF, Wild CP, Montesano R, Willey JC andMann DL. (1989). Carcinogenesis, 10, 2149 – 2153.

Witschi H. (2000). Exp. Lung Res., 26, 743 – 755.World Health Organization. (1997). Tobacco or Health: AGlobal Status Report. Geneva: WHO, pp. 10 – 48.

Yoon JH, Smith LE, Feng Z, Tang M, Lee CS and PfeiferGP. (2001). Cancer Res., 61, 7110 – 7117.

Zhang Y, Yuan F, Wu X, Rechkoblit O, Taylor JS,Geacintov NE and Wang Z. (2000). Nucleic Acids Res.,28, 4717 – 4724.

Zhao C, Tyndyk M, Eide I and Hemminki K. (1999).Mutation Res., 424, 117 – 125.

Zhao C, Vodicka P, Sram R and Hemminki K. (2000).Carcinogenesis, 21, 107 – 111.


7451

Oncogene

Tobacco smoke carcinogens, DNA damage and p53 …

Documents

Transcript of Tobacco smoke carcinogens, DNA damage and p53 …