Anais da Academia Brasileira de Ciências (2007) 79(4): 593-616 [603991]

Anais da Academia Brasileira de Ciências (2007) 79(4): 593-616
(Annals of the Brazilian Academy of Sciences)ISSN 0001-3765www.scielo.br/aabc
Chemical carcinogenesis
PAULA A. OLIVEIRA1, AURA COLAÇO1, RAQUEL CHA VES2, HENRIQUE GUEDES-PINTO2,
LUIS F. DE-LA-CRUZ P.3and CARLOS LOPES4,5
1Department of Veterinary Sciences, CECA V, University of Trás-os-Montes and Alto Douro
5000-801 Vila Real, Portugal
2Center of Genetics and Biotechnology-CGB, University of Trás-os-Montes and Alto Douro (UTAD)
Department of Genetics and Biotechnology, 5000-801 Vila Real, Portugal
3Deparment of Physiology, Faculty of Veterinary, Santiago University, Granxa Street
Campus Universitario, 27002 Lugo, Spain
4Department of Pathology, Portuguese Institute of Oncology, Rua Dr. António Bernardino de Almeida
4200-072 Porto, Portugal
5Departament of Pathology and Molecular Immunology, Institute of Biomedical Sciences Abel Salazar
University of Porto, Largo Professor Abel Salazar, 2, 4099-003 Porto, Portugal
Manuscript received on December 1, 2005; accepted for publication on May 10, 2007;
presented by LUCIA MENDONÇA PREVIATO
ABSTRACT
The use of chemical compounds benefits society in a number of ways. Pesticides, for instance, enable foodstuffs to
be produced in sufficient quantities to satisfy the needs of millions of people, a condition that has led to an increasein levels of life expectancy. Yet, at times, these benefits are offset by certain disadvantages, notably the toxic sideeffects of the chemical compounds used. Exposure to these compounds can have varying effects, ranging from instantdeath to a gradual process of chemical carcinogenesis. There are three stages involved in chemical carcinogenesis.These are defined as initiation, promotion and progression. Each of these stages is characterised by morphological andbiochemical modifications and result from genetic and/or epigenetic alterations. These genetic modifications include:mutations in genes that control cell proliferation, cell death and DNA repair – i.e. mutations in proto-oncogenesand tumour suppressing genes. The epigenetic factors, also considered as being non-genetic in character, can alsocontribute to carcinogenesis via epigenetic mechanisms which silence gene expression. The control of responsesto carcinogenesis through the application of several chemical, biochemical and biological techniques facilitates theidentification of those basic mechanisms involved in neoplasic development. Experimental assays with laboratoryanimals, epidemiological studies and quick tests enable the identification of carcinogenic compounds, the dissection ofmany aspects of carcinogenesis, and the establishment of effective strategies to prevent the cancer which results fromexposure to chemicals.
Key words: cancer stages,carcinogenesis evaluation, chemical carcinogens, chemical carcinogenesis.
INTRODUCTION
Public opinion considers cancer to be an increasingly
threatening disease, affecting people of all ages. After
cardiovascular diseases, it is the second cause of death
amongst the global population (Huff 1994, Weisburger
Correspondence to: Paula A. Oliveira
E-mail: pamo@utad.pt1999). People tend to accept cancer with stoicism and
submit themselves to prolonged periods of treatments,
which are not always effective (Weisburger 1999). The
word carcinogenic was defined as the capacity of a com-
pound to unchain the process of cancer development
in man and animals under the appropriate conditions,
by acting on one of several organs or tissues (Gomes-
An Acad Bras Cienc (2007) 79(4)

594 PAULA A. OLIVEIRA et al.
Carneiro et al. 1997, Huff 1999). With the discovery
of different mechanisms involved in carcinogenesis, this
definition is now incomplete (Butterworth and Bogdanf-
fy 1999). From an experimental point of view, a com-
pound is considered carcinogenic when its administra-tion to laboratory animals induces a statistically signifi-
cant rise in the incidence of one or more histological types
of neoplasia, compared with the animals in the control
group which are not exposed to the substance (Gutiérrez
and Salsamendi 2001).
The factors responsible for cancer development are
classified as exogenous and endogenous (Camargo etal. 1999, Gutiérrez and Salsamendi 2001). The firstgroup includes nutritional habits (food preservation andpreparation), socio-economic status, lifestyle, physicalagents (ionising and non-ionising radiation), chemicalcompounds (natural and synthetic) and biological agents(Helicobacter pylori , Epstein Barr virus, human T lym-
photropic viruses I and II, human papilloma virus and thehepatitis B virus, parasites such as Schistosoma haemo-
tobium ,Clonorchis sinensis andOpisthorchis vivarium ;
growth factors) (Pitot and Dragan 1991, Barrett and An-derson 1993, Farmer 1994, Weisburger 1999, Minamotoet al. 2000, Lutz 2002). Unhealthy lifestyle habits suchas: excess alcohol consumption; inhalation of tobaccoand related products; the ingestion of certain foods andtheir contamination by mycotoxins; are responsible forhigher incidences of certain types of neoplasias in a num-ber of population groups (Gomes-Carneiro et al. 1997,Weisburger 1999, Gutiérrez and Salsamendi 2001). En-dogenous factors include immune system damage andinflammation caused by uncertain aetiology (e.g. ulcer-ative colitis, pancreatitis, etc.), genetic makeup, age, en-docrine balance and physiological condition (Cohen etal. 1991, Barrett and Anderson 1993, Huff 1994, Koivu-salo et al. 1994, Weisburger 1999, Minamoto et al. 2000,Gutiérrez and Salsamendi 2001, Dewhirst et al. 2003,Ohshima et al. 2003, 2005).
Epidemiological studies of cancer incidence de-
monstrated that the risk of developing cancer varies be-
tween population groups and these differences are as-
sociated with lifestyle factors and habits (Garner 1998,
Lai and Shields 1999, Gutiérrez and Salsamendi 2001).
Population migration has resulted in the development of
types of cancer typical of particular geographical areas(King et al. 1995, Gutiérrez and Salsamendi 2001).
The relationship between chemical substances in
the workplace and the development of certain neoplasias
in various occupational groups led to the conception of
experimental models to better understand the biopatho-
logical processes inherent to carcinogenesis (Weinstein1991, Cohen et al. 1992, Gutiérrez and Salsamendi
2001).
Boveri laid down the genetic basis of neoplasic de-
velopment for the first time in 1914 with his theory of
somatic mutation in cancer cells. However at the time,experts in the area of chemical carcinogenesis attributedlittle importance to this hypothesis, considering it to be
pure speculation, instead choosing to put their faith in the
lesser knowledge already available (Weisburger 1999).Between 1980 and 1990, the discoveries made via themolecular biology of proto-oncogenes and tumour sup-pressor genes strengthened the case behind this suppo-sition (Cohen 1998). Neoplasic development bases it-self on the existence of several genetic mutations, de-spite the number not being known. In most of the casesit is assumed to vary between tissues and between dif-ferent species (Grisham et al. 1984, Cohen 1995, 1998,Simons 1995, van Leeuwen and Zonneveld 2001, Lutz2001, Gutiérrez and Salsamendi 2001). During cell divi-sion, spontaneous genetic errors occur. It is estimated tohappen at a frequency of around 10
−5to 10−6through nu-
cleotides and cell division. If the damage reaches a gene
responsible for neoplasic development then the probabil-
ity of developing cancer will be greater (Cohen 1995).
A cancer is made up of billions of cells, all originat-
ing from an initial cell which multiplies clonally, escapes
to apoptosis and accumulates genetic (and/or epigenetic)
alterations which converge into a neoplasic cell (Trosko
2001). The blocking of apoptosis in the face of sig-
nificant genetic damage can ease the accumulation ofaberrant cells and it can become a critical point in malig-nance pathogenesis (Nguyen-ba and Vasseur 1999, Quet al. 2002).
Neoplasias can be classified as benign or malign
depending on their cellular characteristics. The consti-tuent cells of a malign neoplasia show yet more changesin cell biology (Fig. 1). They proliferate autonomously,
differentiate themselves, invade adjacent tissues and fre-
quently metastasize on tissues that are not related to the
An Acad Bras Cienc (2007) 79(4)

CHEMICAL CARCINOGENESIS 595
Limitless replicative
potential Insensitivity to anti-growth
signals
Tissue invasion and
metastasis Neoplasic
differentiation
Evading apoptosisSelf-sufficiency in
growth signals
Telomerase
expression
Angiogenesis
sustained Neoplasic cell
Fig. 1 – Malignant cell characteristics.
primary neoplasia (Hanahan and Weinberg 2000, Shac-
ter and Weitzman 2002). Cells, which are part of benignneoplasias, grow more slowly, and in general, they do notdisturb normal tissue function, unless they compress vitalstructures (Player et al. 2004). The histopathological ob-servation of neoplasias, be they induced or spontaneous,enables us to better evaluate carcinogenesis, but it maynot be enough to identify more subtle alterations such as
molecular changes (Huff 1992, Maronpot 1996).
This review aims to describe of different events in-
volved in chemical carcinogenesis. So, our work starts
with a historical perspective of the study of chemical
carcinogenesis; we will describe the different stages in-
volved in carcinogenesis; the absorption and metabol-
ism of chemical carcinogens. We will classify differenttypes of carcinogens in function of their active mecha-nisms and we will describe the molecular targets of car-
cinogens. Finally, we will describe a selection of the
methods available for evaluating the carcinogenic poten-
tial of chemical compounds.
HISTORICAL PERSPECTIVE OF CHEMICAL
CARCINOGENESIS STUDY
Cancer was described for the first time by Hippocrates
as ‘karkinos’. Galeno introduced the word neoplasia only
in the II century; he defined it as the growth of a bodyarea adverse to nature (Gutiérrez and Salsamendi 2001).Edwin Smith’s papyruses, dating from the XVII century,
describe breast tumefaction.
According to Hayes (1995), it was the English sur-
geon Percivall Pott who first recognized in 1775 thecasual relationship between exposure to environmentalsubstances and neoplasic development. This author de-scribed the occurrence of cancerous alterations in the
skin of the scrotum of London chimney sweeps as a
consequence of repeated localised contamination with
soot. Some years later, and based on these observa-
tions, a guide distributed to Danish chimney sweeps rec-
ommended that these professionals take a daily bath to
avoid such an occurrence (Hayes 1995, Gutiérrez and
Salsamendi 2001). Still in the XVIII century John Hill
observed a high proportion of nasal mucosa cancer inhis patients, and traced it to the localised long-term ex-posure to snuff. In 1890, a high incidence of bladder
cancer in chemical and rubber industry workers was ob-
served across Europe. (Cohen and Ellwein 1991, Gomes-
Carneiro et al. 1997, Garner 1998, Dybdahl et al. 1999,Huff 1999, Bertram 2001). By the end of the nine-teenth century it had become evident that occupational
exposure to certain chemicals or mixtures of chemicals
had carcinogenic effects (Luch 2005). The all-important
next step was to systematically investigate and repro-
duce these diseases in experimental surroundings. The
first experimental work on chemical carcinogenesis was
An Acad Bras Cienc (2007) 79(4)

596 PAULA A. OLIVEIRA et al.
carried out in 1915 by the pathologist Katsusaburo Yama-
giwa and his assistant Koichi Ichikawa (Yamagiwa and
Ichikawa 1918). They rubbed rabbit ears with coal tar
and observed the development of papillomas and carcino-
mas. Meanwhile, others researchers studied carcinogen-esis of the bladder, liver, kidney, pancreas and lung using
laboratory animals. Its success laid the foundations of
the experimental use of animals in the study of human
diseases (Toth 2001). Later, Beremblum and Shubik
used polycyclic aromatic hydrocarbons and croton oil to
study skin carcinogenesis in mice and demonstrate thatcancer development includes several stages (Beremblumand Shubik 1947). When applied in low doses, none ofthese substances have carcinogenic properties by them-selves. Yet, when mixed and in equal doses, they in-duced neoplasic development. The order of expositionto these substances was fundamental for carcinogenesis.Neoplasias developed only when the hydrocarbons wereused first and then the croton oil, never the other wayaround. These authors felt that the carcinogenic actionof these substances was responsible for converting nor-mal cells into neoplasic cells. For them, carcinogenesiswas a complex process including one phase called initia-tion and another called promotion, with one or more ge-netic changes necessary for cancer development. Duringthe next decade, Foulds (1954) introduced the term pro-gression by studying breast adenocarcinoma in female
mice. In the pre-Watson and Crick era, before carcino-
gens were known to bind to DNA, the cancers producedby chemical carcinogens were believed to be due to theirinteraction with proteins in specific tissues (Miller andMiller 1952). By the end of the 1960s, increasing evi-dence pointed to a correlation between the DNA bind-ing capacity of a particular carcinogen and its biologicalpotency (Luch 2005).
STAGES OF CARCINOGENESIS
Studies conducted using animal models, “in vitro” stud-
ies and epidemiologic assays enabled investigators to
conclude that neoplasic pathogenesis is a complex pro-
cess which can be divided into three distinct stages, froman operational point of view. These are: initiation, pro-motion and progression (Foulds 1954, Grisham et al.
1984, Cohen 1991, Mehta 1995, Hasegawa et al. 1998,Gutiérrez and Salsamendi 2001, Trosko 2001).Changes in the genome’s structure occur across the
three stages of neoplasic development (Simons 1995,
Pitot 2001, Luch 2005). Changes in gene expression
also take place during the promotion stage, with selec-
tive proliferation of initiated cells and the development
of pre-neoplastic cells (Grisham et al. 1984, Gutiérrezand Salsamendi 2001). During initiation and promo-tion, apoptosis and cell proliferation can occur at differ-ent rates, while remaining balanced. During progression,this balance is modified and from there malignancy arises(Mehta 1995) (Fig. 2).
Human life is led under very different conditions
from these experimental procedures. Although the pro-
cess of carcinogenesis is similar for man and experimen-
tal animals, the different chemical compounds to which
humans are exposed throughout their lives alter the speed
of the process and the frequency of mutation, the speedof cell growth and the phenotypical expression of the
changed genes. On the other hand, the individual’s sus-ceptibility and their defence mechanisms have their owninteraction, which modifies each of the neoplasic stages.
INITIATION
The first stage of carcinogenesis has been labelled ini-tiation since 1947 (Beremblum and Shubik 1947). Theconclusions reached from several experiments enabledthe conclusion to be drawn that initiation is caused byirreversible genetic changes which predispose suscep-tible normal cells to malign evolution and immortality(Beremblum and Shubik 1947, Stenbäck et al. 1981,Butterworth et al. 1992, Mehta 1995, Dybing and Sanner1999, Trosko 2001, 2003, Shacter and Weitzman 2002).The initiated cell is not a neoplasic cell but has taken itsfirst step towards this state, after successive genotypicaland phenotypical changes have occurred (Trosko 2003).From a phenotypical perspective, the initiated cell is sim-ilar to the remaining cells. It undergoes mutations andthese induce proliferation but not differentiation (Trosko2001).
DNA damage has been well established as the event
which kick-starts chemical carcinogenesis (Santella et
al. 2005). DNA damage can be repaired by enzymatic
mechanisms (Bertram 2001, Jeng et al. 2001, Shacter
and Weitzman 2002). Cells which are proliferating have
less time to repair the damaged DNA and remove co-
An Acad Bras Cienc (2007) 79(4)

CHEMICAL CARCINOGENESIS 597
CCHHEEMMIICCAALLSS
INITIATION PROMOTION
Cells with adductsInitiated cells Normal cells DNA repair Cell proliferation Cellular
proliferation
PROGRESSION
CANCER APOPTOSIS
CELL TOXICITYCell
proliferation
Fig. 2 – Chemical carcinogenesis stages and the occurrences involved in each one.
valent bonds that chemicals establish with the DNA –
known as adducts (Heidelberger 1977, Richardson et al.1986, Frowein 2000).
At this stage, the initiated cells can remain latent
for weeks, months or years, or they can grow in an auto-
nomous and clonal fashion (Scott et al. 1984, Dybing
and Sanner 1999, Player et al. 2004). This initiationprocess ensures that cellular division remains symmetri-cal by creating two new initiated cells (Trosko 2003).
The clonal expansion of initiated cells results from a
mitogenic process caused by an increase in the number
of new cells and apoptosis inhibition, which prevents ini-tiated cells from dying off (Trosko 2001).
The increase in DNA damage is specifically impor-
tant to stem cells, because they survive for a long time
and exist in several tissues (Potter 1978, Simons 1999,
Trosko 2001, Williams 2001). In 1978, Potter explained
that neoplasic cells could display a phenotype established
between the embryonic aspect and the terminal differen-
tiation, and that all neoplasic cells had monoclonal origin
from a stem cell. By definition, stem cells are immortalcells until they differentiate, or death is induced. If wedelay their differentiation they become initiated and ac-cumulate in tissues as clones of abnormal cells (Trosko2003). Although stem cells are not identifiable in most
tissues, it is believed that every tissue has a populationof stem cells (Player et al. 2004).
Initiation is a fast, irreversible phenomenon and istransmitted to daughter cells (Farber 1984). Cell pro-
liferation is essential for this stage, if cellular divisionoccurs before DNA repair systems can act then the in-jury becomes permanent and irreversible. Initiation is anadditive process, neoplasic development depends on thecarcinogenic dose, increasing the dose increases the in-
cidence and the multiplicity of resultant neoplasias and
reduces the latent period of its manifestation. Not all
cells of a living organism exposed to an initiator agent
will be initiated even if they have suffered mutations, and
the genes that regulate the terminal differentiation mustalso be mutated (Farber 1984, Yuspa and Poirier 1988,Klaunig et al. 2000, Trosko 2001).
Spontaneously initiated cells exist in all living or-
ganisms (Gomes-Carneiro et al. 1997, Trosko 2001). Ini-tiation can begin with spontaneous mutations, supportedby normal occurrences such as DNA depurination and
deamination. Errors in DNA replication are also asso-
ciated with initiation. Although spontaneous initiation
is less common than induced initiation, its existence hasbeen confirmed by the occurrence of spontaneous neo-plasias in laboratory animals (Pitot and Dragan 1991,Gomes-Carneiro et al. 1997).
PROMOTION
The concept of promotion was introduced when chemi-
cal substances with low carcinogenic activity were dis-covered, which were still able to induce the develop-
An Acad Bras Cienc (2007) 79(4)

598 PAULA A. OLIVEIRA et al.
ment of cancer under experimental conditions (Berem-
blum and Shubik 1947).
Promoter compounds do not interact directly with
DNA and unchain biological effects without being meta-
bolically activated (Yuspa et al. 1983, Butterworth et al.1992, Weisburger 1998, Williams 2001). These agents
increase cell proliferation in susceptible tissues, con-
tribute towards fixing mutations, enhance alterations in
genetic expression and cause changes in cellular growth
control (Mehta 1995, Gomes-Carneiro et al. 1997). On
the other hand, these promoters may indirectly damageDNA by oxidation (Gutiérrez and Salsamendi 2001).At first, these occurrences were associated with epige-netic mechanisms, but nowadays it is widely agreed thatpromotion also involves genetic changes (Simons 1995,Hanahan and Weinberg 2000).
Promoters delay the natural inhibition of the quies-
cent cells or in G
0by gap junctions (Barrett and Ander-
son 1993, Simons 1999, Bertram 2001, Trosko 2001).The promoters’ most important activity is mitogenesis– genotoxical and mutational actions are not necessaryat this stage (Pitot and Dragan 1991). The promotermust be present for weeks, months and years in order
to be effective and its effectiveness depends on its con-
centration in the target tissue (Butterworth et al. 1992).
Promotion is a reversible stage, after a promoter’s dis-
appearance a regression in cell proliferation can occur,probably by apoptosis. It is a stage that can be mouldedup by physiological factors and therefore limit the extentof experimental carcinogenesis. Some promoter agentsare specific for a particular tissue, but others act simul-
taneously upon several tissues (Yuspa et al. 1983, Scott
et al. 1984, Yuspa and Poirier 1988, Gutiérrez and Sal-
samendi 2001).
In studies of chemical carcinogenesis with pro-
longed exposure and using high doses almost all of the
promoter agents induce neoplasias without initiation(Pitot and Dragan 1991, Gutiérrez and Salsamendi 2001).Exposure to phenobarbital, benzene, asbestos, and ar-
senic even without the previous application of initiator
agents leads to neoplasic development (Melnick et al.
1996, Trosko 2001). This contradiction has two possibleexplanations: either the genotoxic effect was not iden-tified by mutagenicity and genotoxicity assays, or the
initiated cells emerged spontaneously. In this last casewe may consider that the promoter has an indirect effect
– by increasing the frequency of cellular division it en-courages the appearance of errors in DNA replication,as well as mutations.
Not all cells exposed to promoters take part in the
promotion stage, only cells which are stimulated to di-
vide, that are undifferentiated, and have survived apop-
tosis, can contribute to instability between growth and
cell death and lead to the appearance of a malign neo-plasia (Trosko 2001).
PROGRESSION
The sequence of lesions identified, via histopathology,between initiation and promotion are designated as pre-
neoplastic lesions and/or benign neoplasias (Gutiérrezand Salsamendi 2001). Their transformation into ma-lign lesions is the last of the stages of carcinogenesis andis the most extended – it is labelled progression (Klauniget al. 2000, Williams 2001). In progression, a neopla-sic phenotype is acquired through genetic and epigeneticmechanisms (Shacter and Weitzman 2002). During pro-gression, cell proliferation is independent from the pres-ence of stimulus (Lutz 2000, Gutiérrez and Salsamendi2001).
Progression is characterised by irreversibility, ge-
netic instability, faster growth, invasion, metastization,and changes in the biochemical, metabolical and mor-phological characteristics of cells (Pitot and Dragan1991, Butterworth et al. 1998, Loeb 1998, Klaunig etal. 2000, Gutiérrez and Salsamendi 2001, Dixon andKopras 2004).
Angiogenesis, as an epigenetic occurrence, is es-
sential to neoplasic progression. The acquisition of an
angiogenic phenotype precedes the development of char-acteristics that contribute to malignancy and its inhibitiondelays neoplasic development (Hawighorst et al. 2001).
ABSORPTION AND METABOLISM OF
CHEMICAL CARCINOGENS
Following exposure, chemical carcinogens may be ab-
sorbed in a number of ways (oral, inhalator, cutaneous,and injection) and distributed across several tissues (Con-noly et al. 1988). Absorption depends on the physico-
chemical properties of the substance and can take placevia passive or active transport. The substances absorbed
An Acad Bras Cienc (2007) 79(4)

CHEMICAL CARCINOGENESIS 599
orally pass through the liver and only then are they dis-
tributed in the body; those absorbed in the lung are dis-
tributed by the blood before reaching the liver at a later
stage (King et al. 1995, van Leeuwen and Zonneveld
2001). Those carcinogenic compounds classified as di-rect act directly on DNA, but most require enzymatic con-
version and are thus labelled as indirect or procarcino-
gens (Sarasin and Meunier-Rotival 1976, Hayes 1995,
Lai and Shields 1999, Klaunig et al. 2000, Oesch et al.
2000, Poirier et al. 2000, Luch 2005). Metabolic ac-
tivation is controlled by phase I reactions, while phaseII reactions protect the body through the transformationof activated compounds into inert products which areeasily eliminated from the body (Fig. 3) (Hayes 1995,Bartsch and Hietanen 1996, Mostafa et al. 1999, Klauniget al. 2000, Gonzalez and Kimura 2001, van Leeuwenand Zonneveld 2001, Park et al. 2005).
The performance of metabolic enzymes is essential
for understanding chemical carcinogenesis and learningthe differences between species as far as their suscep-tibility to neoplasic development is concerned (Sarasinand Meunier-Rotival 1976, Lai and Shields 1999, Guen-guerish 2000, 2001, Gonzalez 2001). The enzymes inphase I participate in the reactions of oxidation, reduc-tion and hydrolysis, and are classified as oxidoreductases(cytochrome P450 dependent monooxygenases, flavinemonooxygenases, cyclooxygenases and alcohol dehy-drogenase) and hydrolases (epoxide hydrolases) (Hayes1995, Garner 1998, Galati et al. 2000, Oesch et al. 2000,Garcea et al. 2003). Phase II enzymes participate in theconjugation and inactivation of chemical carcinogensand include transferases (glutathione S-transferases, N-acetyltransferases, UDP-glucuronosyltransferases, sul-photransferases) (Oesch et al. 2000, Guengerich 2000,Gonzalez 2001). Although these enzymes were origi-nally only thought to be involved in the detoxificationstages of biotransformation, they can also contribute tothe activation of certain procarcinogens in vivo (Luch
2005).
Metabolic activation occurs predominantly in the
liver at the plain endoplasmic reticulum where the cy-
tochrome P450 is more abundant, and to a lesser degree
in the bladder, skin, gastrointestinal system, oesopha-
gus, kidneys, and lungs (Bartsch and Hietanen 1996,
Mostafa et al. 1999, Guengerich 2001, van Leeuwen andZonneveld 2001, Oda 2004). During this phase the cy-
tochrome P450 mono-oxygenases introduces a reactive
polar group into the carcinogenic, making it lipophylic.
It then converts it into a powerful electrophilic prod-
uct capable of establishing adducts with DNA (Straub
and Burlingame 1981, Lai and Shields 1999, Galati etal. 2000, Park et al. 2005). Phase II reactions are catal-
ysed by hepatic and extra hepatic, cytoplasmic and cy-
tochromic enzymes, acting separately or joined together
(Gonzalez 2001). Conjugation reactions enable these en-
zymes to decompose the polar group in glucose, aminoacids, glutathione and sulphate, which are less toxicmetabolites that are more soluble in water and more eas-
ily expelled by the urine and bile (Galati et al. 2000,
Oesch et al. 2000, Gonzalez and Kimura 2001, vanLeeuwen and Zonneveld 2001).
Peroxidations also occur parallel to metabolic reac-
tions with the continuous production of reactive oxygenspecies (ROS) (Weisburger 1999, Klaunig et al. 2000,Ohshima et al. 2005). These radicals are associated withseveral chronic diseases including chemical carcinogen-esis (Klaunig et al. 2000). The ROS damage DNA, RNA,and proteins by chemical reactions such as oxidation, ni-tration/nitrosation and halogenation. This leads to anincrease in mutations and alterations in the functions ofimportant enzymes and proteins (Park et al. 2005). Sev-eral experiments have proved that chemical compounds,which create ROS in excess, encourage initiation, pro-motion and neoplasic progression through genotoxicity(Galati et al 2000, Shacter and Weitzman 2002). Theimpact of the ROS controlled by a cellular mechanismthat operates at different levels: metabolism; reactionsthat maintain the redox balance in cells; transductionof the signal regulator of oxidation and DNA reparation(Bolt et al. 2004).
Park et al. (2005) says that the same enzyme may
have the capacity to activate one chemical and deacti-vate another, all depending on its chemical structure.The specificity of the activation systems of different tis-sues regulate neoplasic development and is dependenton genetic polymorphism, which requires the expres-
sion and distribution of the enzymes involved in phase I
and II reactions, and the resulting susceptibility to can-
cer development (Schut and Castonguay 1984, Hayes
1995, Henglster et al. 1998, Mostafa et al. 1999, Dybing
An Acad Bras Cienc (2007) 79(4)

600 PAULA A. OLIVEIRA et al.
CARCINOGENIC EXPOSITION
ABSORPTION
DISTRIBUTION
ACTIVATION INACTIVATION
EXCRETION
(Kidneys, liver, lungs) GENOTOXIC MECHANISMS
•DNA adducts
Chromosome breakage, fusion,
deletion, mis-segregation, non disjunction NON-GENOTOXIC MECHANISMS
Inflammation
Immunosupression
Reactive oxygen species
Reactive nitrogen species
Receptor activation
Epigenetic silencing
Genomic damage Altered signal transduction
Hypermutability
Genetic instability
Loss of proliferation control
Resistance to apoptosis Cancer BIOTRANSFORMATION
(LIVER, KIDNEYS, LUNGS)
Fig. 3 – Metabolic activation of chemical compounds and genotoxic and non-genotoxic effects of carcinogens.
and Sanner 1999, Gonzalez 2001, Gonzalez and Kimura
2001, Gutiérrez and Salsamendi 2001, Lutz 2002). Peo-ple with a high quantity of phase I and a low quantity ofphase II enzymes have a higher probability of synthesis-ing intermediate compounds and exhibiting more DNAdamage (Rojas et al. 2000).
The previously described metabolic methods areequally important for both humans and animals, although
there exist qualitative and quantitative differences be-
tween them. These have lead to incorrect interpreta-tions when animal models are used in the research andanalysis of carcinogenic properties of chemical com-
pounds (Guengerich 2000, Gonzalez 2001, Gonzalez and
Kimura 2001).
An Acad Bras Cienc (2007) 79(4)

CHEMICAL CARCINOGENESIS 601
Several studies have been developed in order to
evaluate the differences between several exogenous and
endogenous factors on individual susceptibility to carci-
nogenesis (Table I) (Barrett 1993, Bartsch and Hietanen
1996, Maronpot 1996, Lutz 1998, 1999, Ishikawa et al.2001, Miller et al. 2001).
TABLE I
Factors that control and change chemical car cinogenesis.
Age V irus
Sex Diet, nutrition and life style
Animal species Genetic constitution
Endocrine system Anticancer drugs
Immune system Metabolic w ays
Trauma DNA repair
Radiation
CARCINOGENIC CLASSIFICATION
Carcinogenic classification is by no means consensual
(Butterworth and Bogdanffy 1999, Bolt et al. 2004). It
is not easy to incorporate a carcinogenic compound intoa certain group because the information obtained fromdifferent studies is increasingly complex (Pitot andDragan 1991, Butterworth et al. 1992). Some authorsclassify them in function of their participation in eachof the stages of carcinogenesis. In this way, incompletecarcinogens are mutagenic chemicals that instigate irre-versible DNA damage (Mirsalis et al. 1990, Pitot andDragan 1991). A complete carcinogen displays pro-perties of both initiators and promoters simultane-ously depending on the dosage and exposure time (Pitotand Dragan 1991, Farmer 1994, Hasegawa et al. 1998,
Trosko 2001).
Other authors classify chemical carcinogens in
function of their mechanisms of action as being geno-toxic and non-genotoxic (mitogenic and cytogenic)(Cohen and Ellwein 1991, Butterworth et al. 1992,Nguyen-ba and Vasseur 1999, Klaunig et al. 2000,Williams 2001). The knowledge about the mechanism
of action of non-genotoxic carcinogens is known to be
inferior to that of genotoxic carcinogens.
Genotoxic carcinogens are complete carcinogens
and qualitatively and quantitatively change a cell’s ge-netic information (Trosko 2001). They exhibit a directanalogy between their structure and activity, are muta-
genic on in vitro assays, are active in high doses, andmay affect several animal species, and damage differentorgans (Klaunig et al. 2000, Gutiérrez and Salsamendi2001, Luch 2005). In high doses, they cause toxicityand cell proliferation, increasing DNA replication andinfluencing its carcinogenic activity (Cohen 1998). Fol-lowing transmembranar diffusion they are metabolizedin electrophilic compounds that enter the nucleus andinteract with nucleophilic sites (DNA, RNA and pro-teins) changing their structural integrity and establish-ing covalent bonds known as adducts (Miller and Miller1975, Straub and Burlingame 1981, Cohen et al. 1992,Ashby 1996, Weisburger 1998, Frowein 2000, Bertram2001, Lutz 2001, Williams 2001, Baird and Mahadevan
2004). The formation of adducts constitutes the first crit-
ical step of carcinogenesis and if these are not repairedbefore DNA replication then mutations may occur in theproto-oncogenes and tumour suppressor genes, whichare essential for the initiation stage (Sobels 1975, Barrettand Wiseman 1987, Farmer 1994, Lutz 2001, Williams
2001, Li et al. 2005). The number of adducts formed
by carcinogens is changeable and each of them maycause a specific damage to DNA (Straub and Burlingame1981, Farmer 1994, Otteneder and Lutz 1999). Mu-tations linked to adducts can appear through deletion,frameshift, or by nucleotide substitution (Garner 1998).Mutations cause an undefined number of cell changes,
translated into aberrant protein expression and in changes
in cell cycle control. Adducts assume importance inchemical carcinogenesis because of the way they changeDNA, possibly inducing an incorrect transcription andcausing mutations of the new DNA chain. The existenceof many adducts can break the DNA chain, causing mu-tation or loss of genetic material (Cohen 1995, Hayesand Pulford 1995, Trosko 2001). Adduct repair is coor-dinated by several enzymes and controlled by differentgenes. It can be done via the excision of bases, or nu-cleotides, recombined repair or mismatch repair (Farmer1994, Moustacchi 1998, Miller et al. 2001, Hanawalt etal. 2003).
The identification of adducts suggests that chemical
carcinogens are absorbed, metabolized and distributed bytissues, thus fleeing from the body’s detoxification andrepair mechanisms (Garner 1998, Airoldi et al. 1999,
An Acad Bras Cienc (2007) 79(4)

602 PAULA A. OLIVEIRA et al.
Guengerich 2000). The identification and analysis of
adducts can be carried out using marked radioactive car-cinogens, those most-commonly used are
14C and tri-
tium, each adduct can be identified by their 106or 107
nucleotides (Garner 1998). However, the most used tech-niques are immunoassays with
32P, gaseous chromatog-
raphy associated with mass spectrometry and HPLC as-sociated with fluorescent spectroscopy (Farmer 1994,Airoldi et al. 1999). There are also monoclonal andpolyclonal antibodies available on the market which areused to identify adducts by immunohistochemistry (San-tella et al. 2005). There is a positive correlation betweenthe quantity of adducts detected in animal models andthe number of neoplasias developed (Yuspa and Poirier1988, Williams 2001, Baird and Mahadevan 2004).
Non-genotoxic carcinogens act as promoters and do
not need metabolical activation. They do not react di-rectly with DNA, do not raise adducts and show negativeon mutagenicity tests carried out in vivo and in vitro(Butterworth et al. 1992, Melnick et al. 1996, Butter-worth and Bogdanffy 1999, Klaunig et al. 2000, Gon-zalez 2001, Williams 2001). These compounds modu-late growth and cell death, potentate the effects of geno-toxic compounds, do not show a direct correlation be-tween structure and activity, and their action is limited bytheir concentration. They are tissue- and species-specific(Farmer 1994, Melnick et al. 1996, Gomes-Carneiro etal. 1997, Butterworth and Bogdanffy 1999, Klaunig etal. 2000). Melnick et al. (1996) states that exposureto these compounds favours the synthesis of other sub-stances responsible for neoplasic development. Thesecompounds promote effects on target cells which indi-rectly unchain neoplasic transformation or increase neo-
plasic development from genetically changed cells (Wil-
liams 2001). Non-genotoxic carcinogens are classifiedas cytotoxic and mitogenic in function of whether theiractivity is mediated by a receptor or not (Cohen 1991,Cohen et al. 1992, Butterworth and Bogdanffy 1999).Mitogenic compounds such as phorbol esters, dioxins,and phenobarbital induce cell proliferation in target tis-sue through interaction with a specific cellular receptor(Cohen et al. 1992). Cytotoxic carcinogens cause cell
death in susceptible tissues followed by compensatoryhyperplasia, taking chloroform as an example (Cohen etal. 1991, Butterworth et al. 1992, Klaunig et al. 2000).
If the carcinogen dose is high, some cells cannot sur-vive. The more that nearby cells increase the numberof cell divisions through regenerative procedures, themore likely it is that they will end up being prematurelyrecruited for the cell cycle and that the time availablefor reparation DNA will be inferior – this increases theprobability of mutations occurring (Cohen 1991, Mel-nick et al. 1996). On the other hand, necrosed cellsare destroyed by the immune system and ROS, reac-tive nitrogen species (RNS), and proteolytic enzymesare produced (Lutz 1998, Ohshima et al. 2005). Whenproduction of these ROS and RNS exceeds the cellularanti-oxidant capacity, it may cause oxidative damages tolipids, proteins, carbohydrates, and nucleic acids, leadingto carcinogenesis and cell death (Ohshima et al. 2005).Mitogenic compounds need to be present in certain con-centrations to promote their activity. Contrastingly, theaction of non-cytotoxic compounds is independent oftheir concentrations (Butterworth et al. 1992, Butter-worth and Bogdanffy 1999).
Chemical carcinogens can be classified into several
groups, on Table II we brought them together under thefollowing headings: Group, compound, mechanism ofaction, and affected organs/cancer type.
As we mentioned before, the classification of the
carcinogenic compounds according to their mechanismof action continues to cause controversy. Bolt et al.
(2004) propose the division of genotoxic compounds
into two groups: those which react with DNA, and geno-toxic at a chromosomal level. Compounds, which reactwith DNA, are subdivided into three different groups:initiators (with unlimited doses), borderline, and weakgenotoxic (they act by secondary mechanisms) (Fig. 4).
Chemical carcinogens can have additional synergic
or antagonistic effects when simultaneously presentedin different metabolic ways (Schmahl 1976, Lutz 2001).The synergy between smoking and exposure to asbestosfavours lung cancer development as a consequence ofchronic inflammation and compensatory cell prolifera-tion. This antagonism may be exemplified by the protec-tive action of fruit and vegetables in the modulation ofindividual susceptibility to neoplasic development (Lutz2001, 2002).
An Acad Bras Cienc (2007) 79(4)

CHEMICAL CARCINOGENESIS 603
TABLE II
Chemical car cinogens.
Group Compound Mechanism of actionAffected org ans/
Cancer type
Polyc yclic Benzo[a]p yrene Form adducts with purine Skin, lungs, stomach
aromatic Polychlorinated biphen yls bases of DNA, mainly Liver skin
hydrocarbons (Luch 2005) resulting on transv ersions
Aromatic 2-Acetylaminofluorene Genotoxic compounds, Liver, bladder
amines/amides 4-Aminobiphen yl increase the rate of Bladder
2-Naphth ylamine cell duplication Bladder
(Luch 2005)
Aminoazo o-Aminoazotoluene Forms adducts with Liver, lungs, bladder
dyes N,N-dimeth yl-4- DNA and with Lungs, liv er
aminoazobenzene haemoglobin
(Golka et al. 2004)
N-nitroso N-Nitrosodimeth ylamine Form adducts at N- and Liver, lungs, kidneys
compounds (Drablos et al. 1998) O-atoms in DNA bases
Carbamates N-methylcarbamate esters Chromosome aberration, Experimental results
(Wang et al. 1998) gene mutation, showed liver, kidne ys
celltransformation and tests de generation
Halogenated Trichloroeth ylene Somatic mutations, Experimental results
compounds (Lock et al. 2007) modification of cell showed kidney, liv er
cycle pathw ays and lung cancer
Natural Aflatoxin B1 Forms adducts with Liver
carcinogens (Wild et al. 1986) guanine, react with Lungs
Asbestos (Luch 2005) RNA and proteins
Metals Arsenic (Shi et al. 2004) Oxidative stress Skin, lungs, liv er
Cadmium (Hartwig et al. 2002) Inhibit DNA repair Lungs, prostate,
pathways and nucleotide- kidne ys
excision repair
Nickel (Costa et al. 2003) Histone acetylation Lungs, nasal ca vity
and DNA hypermeth ylation
Anticancer Alkylating agents Interstrand and/or Leukaemia
drugs (Luch 2005) intrastrand cross-links
EPIGENETIC MECHANISMS INVOLVED IN CHEMICAL
CARCINOGENESIS
The most well understood epigenetic mechanisms in-
volve DNA methylation and histone acetylation, methy-
lation, and phosphorylation (Fig. 5). Demethylation of
promoter regions at the CpG sequences can lead to an
over-expression of proto-oncogenes, and silencing ofgene expression can occur as a result of hypermethy-lation, sometimes leading to chromosome condensation
(Klaunig et al. 2000). There appears to be a relationshipbetween DNA methylation and histone modifications;patterns of histone deacetylation and histone methyla-
tion are associated with DNA methylation and gene si-lencing. Interestingly, these epigenetic changes in chro-matin can also alter the sensitivity of DNA sequences tomutation, thus rendering genes more susceptible to toxicinsult (Dixon and Kopras 2004).
MOLECULAR TARGETS OF CHEMICAL CARCINOGENS
The discovery of the ability of oncogenes to induce neo-plasic transformation when transfected into immortal-ized mouse cell lines, initially seemed to answer many
An Acad Bras Cienc (2007) 79(4)

604 PAULA A. OLIVEIRA et al.
CCHHEEMMIICCAALLCCAARRCCIINNOOGGEENN
GGEENNOOTTOOXXIICC NNOONN–GGEENNOOTTOOXXIICC
DNA reactive Chromosomic action
Weak genotoxic
BorderlineInitiator
Fig. 4 – New proposal to classify chemical carcinogens.
basic molecular questions about the molecular origins
of cancer. However, it soon became clear that this wasnot the whole picture and that there existed other genesthat could influence neoplasic transformation (Bertram2001). There are several genes which intervene in car-cinogenesis – their identification revolutionised chem-ical carcinogenesis and oncology (Kinzler and V ogel-stein 1997, Bertram 2001). Out of all of these, proto-oncogenes, tumour suppressor genes and cell cycle regu-lator genes assume a particular importance (Mehta 1995,Nguyen-ba and Vasseur 1999, Klaunig et al. 2000). Un-like diseases such as cystic fibrosis or muscular dystro-phy, wherein mutations in one gene can cause disease,no single gene defect “causes” cancer. Mammalian cellshave multiple safeguards to protect them against poten-tially lethal effects of cancer gene mutations, and onlywhen several genes are defective does an invasive can-cer develop. Thus it is best to think of mutated cancergenes as contributing to, rather than causing, cancer (Vo-gelstein and Kinzler 2004). Neoplasic development re-quires errors in cellular defence mechanisms, which arecontrolled by checkpoints that may forbid the entry ofcells with DNA damage into the cell cycle before DNAreparation occurs (blocked at G
1) and the cell divides
(blocked at G 2) (Fig. 6) (Khan et al. 1999, Khan and
Dipple 2000). The capacity of cells to evade the cel-
lular defence mechanism has an undoubted contributiontowards the carcinogenesis (Khan and Dipple 2000).
The tumour suppressor proteins p53; p21 and pRb
play crucial roles in cellular protection, because they en-courage the blocking of cells at G
1(Khan et al. 1999).
The loss of pRb protein function provokes an increase
in the cell proliferation rate and an absence of termi-
nal differentiation. p53 can interrupt the cell cycle atG
1and go on to repair DNA damage (Melnick et al.
1993, Loeb 1998, Khan and Dipple 2000, Pritchard etal. 2003, Dixon and Kopras 2004). The most promi-nent and best-studied tumour suppressor is p53, if DNAis damaged then p53 can induce apoptosis in order tomaintain the stability of the cells’ genome (Klaunig et al.2000, Hanawalt et al. 2003, Babenko et al. 2006). Theloss of p53 during carcinogenesis can predispose pre-neoplastic cells to accumulate additional mutations byblocking the normal apoptotic response to genetic dam-ages (Klaunig et al. 2000). The loss of p53 functionactivates proto-oncogenes and inactivates tumour sup-pressor genes therefore performing an exceptional rolein chemical carcinogenesis (Luch 2005). The biologicalactivity of p53 protein is dependent on its ability to bindtranscriptional regulatory elements in DNA. The searchfor critical genes regulated by p53 led to the discovery ofthe p21 gene. p21 acts as an inhibitor of cyclin-dependentkinases providing a functional link between p53 and cellcycle (Bertram 2001).
An Acad Bras Cienc (2007) 79(4)

CHEMICAL CARCINOGENESIS 605
Fig. 5 – Epigenetic mechanisms involved in chemical carcinogenesis.
S
G1G2 M
STOP G0
Repairs
ahead
DNA repair
genes
Tumour suppressor
genesOncogenes
Fig. 6 – Cell cycle and its control by molecular targets (oncogenes and tumour suppressor genes). The cell cycle is a critical process that a cell
undergoes in order to copy itself exactly. Most cancers cause mutations in the signals that regulate a cell’s cycle of growth and division, namely
in oncogenes (which act as dominant mutations) and tumour suppressor genes (that function recessively). Normal cell division is required for
the generation of new cells during development and for the replacement of old cells as they die. In normal cells, tumour suppressor genes act
as braking signals during G1 to stop or slow the cell cycle before it reaches the S phase. DNA repair genes are active throughout the cell cycle,
particularly during G2 after DNA replication and before the chromosomes prepare for mitosis.
An Acad Bras Cienc (2007) 79(4)

606 PAULA A. OLIVEIRA et al.
A common feature of all the known genetic cancer
syndromes is that they are predisposed only to selective
types of malignancy. However, many of the genes mu-
tated in these syndromes are ubiquitously expressed, and
influence seemingly universal processes such as DNArepair or cell cycle control (Chao and Lipkin 2006).
DNA repair is a process which enables a cell to main-
tain its genome fidelity. There are several routes towards
DNA repair. For example, there is excision repair, which
consists of both nucleotide excision repair (NER) and
base excision repair (BER), mismatch repair (MMR), anddouble strand break (DSB) repair, as reviewed by Fried-berg (2003). Each pathway utilizes unique enzymaticmechanism. In this review we outline the DNA repairprocesses mediated by p53 family target genes (Fig. 7)once the p53 has been mutated in a very large fractionof tumours from nearly every possible source. In theirrole as genomic protectors, it is not surprising that thep53 family have a part to play in DNA repair (Fig. 7).The p53 family participate in NER by inducing the ex-pression of GADD45, xeroderma pigmentosum group Egene [XPE] and XPC (Hwang et al. 1999, Tan and Chu2002, Adimoolam and Ford 2002). GADD45 has alsobeen shown to interact with the core histones and facil-itate topoisomerase relaxing of chromatin (Carrier et al.1999). Defective NER is associated with xeroderma pig-mentosum (XP), an autosomal recessive disorder charac-
terized by excessive skin cancers caused by an extreme
sensitivity to UV light (Harms et al. 2004).
The mismatch repair pathway is also influenced by
the p53 family. p53 and p73 induce the expression of
p53R2, a gene which is homologous with the R2 regula-
tory subunit of ribonucleotide reductase (RNR) (Nakano
et al. 2000). p53R2 functions in a non-specific manner
to increase the pool of free dNTPs when the need for
repair arises. Although p53R2 and R2 are similar, they
differ in their N-terminal amino acid sequence and regu-
lation. p53R2 is induced by p53 and p73, while R2 syn-thesis occurs during S phase. The p53R2 and R1 com-plex functions as an active RNR (Guittet et al. 2001).
p53 upregulates two very important proteins along the
MMR pathway: human MutS homologue 2 (hMSH2)
and proliferating cell nuclear antigen (PCNA) (Schereret al. 2000, Xu and Morris 1999). Mutations of hMSH2result in hereditary nonpolyposis colorectal cancer, a col-orectal cancer syndrome. hMSH2 functions in mismatch
recognition and binds mismatched bases (Lamers et al.
2000). PCNA, a cofactor for DNA polymerase δ,i sa n –
other p53 target gene and has been shown to interact
with hMSH2 to facilitate hMSH2 transfer to mismatched
bases (Flores-Rozas et al. 2000).
Alterations in the ras gene have been identified in
several neoplasias that have been chemically induced in
rodents. Mutations of the ras gene exist in about 20%
of human neoplasias located in the colon, breast, lung,
and bladder (Pritchard et al. 2003). Analysis of the rasgene isolated from the DNA of these neoplasias revealsthat changes in the sequence of nucleotides correspond to
the places where carcinogens interact with DNA. Each
chemical compound creates its own unique fingerprinton DNA (Robbins and Cotran 2005).
Some authors classify the genes involved in car-
cinogenesis as caretaker and gatekeeper (Kinzler andV ogelstein 1997, Lai and Shields 1999). This classi-fication is based on their involvement in maintaininggenome integrity and DNA repair, respectively (Lai andShields 1999). The caretakers are responsible for main-tenance of genome stability. Mutations in the caretakergenes, which are considered to be typical tumour sup-pressors, compromise genome stability and, more specif-ically, increase the probability of mutation in the gate-keepers which include both tumours suppressor genesand oncogenes (Vogelstein and Kinzler 2004, Blagos-klonny 2005). Gatekeeper genes regulate neoplasic de-velopment by inhibiting its growth or killing it (Kinzlerand V ogelstein 1997). In contrast, inactivity by caretakergenes does not support the starting phase of a neoplasia,instead favouring the genetic instability which results inan increase in mutations across all genes, including thegatekeeper. A neoplasia initiated by the inactivity of agatekeeper gene can progress quickly as a consequenceof its effect on genes that directly control cell death(Kinzler and Volgestein 1997).
EVALUATION OF CARCINOGENICITY
A major change in the field of carcinogenesis researchhas occurred over the last two decades with the develop-ment of analytical methods that are sensitive enough todetect background damage to DNA in healthy humans(Sharma and Farmer 2004). The control of responses to
An Acad Bras Cienc (2007) 79(4)

CHEMICAL CARCINOGENESIS 607
p53 degradation in
proteasomasDNA
p21 gene
p21 mRNA
p21
p21p53Mdm2Mdm2
p53
p53NUCLEUS
Active p53
p53 active binds to regulatory
region of p21 gene .
Transcription
Translation
p21 (Cdk inhibitor
protein)
CdkCyclin
CdkCyclin
ACTIVE
G1/S-Cdk and
S-CdkINACTIVE
G1/S-Cdk and
S-Cdk complexed
with p21STOP
Cell
CycleProtein kinase activation and
phosphorilation
of p53
p53 family
NERMMR
XPC
XPE
GADD45p53R2
PCNA
hMSH2
p53 family targetp53 family target
p53 target
p53 targetp53 target
p53 target
p53 degradation in
proteasomasDNA
p21 gene
p21 mRNA
p21
p21p53Mdm2Mdm2Mdm2
p53p53
p53p53NUCLEUS
Active p53
p53 active binds to regulatory
region of p21 gene .
Transcription
Translation
p21 (Cdk inhibitor
protein)
CdkCyclin
CdkCyclin
CdkCyclin
CdkCyclin
ACTIVE
G1/S-Cdk and
S-CdkINACTIVE
G1/S-Cdk and
S-Cdk complexed
with p21STOP
Cell
CycleProtein kinase activation and
phosphorilation
of p53
p53 family
NERMMR
XPC
XPE
GADD45p53R2
PCNA
hMSH2
p53 familyp53 family
NERMMR
XPCXPC
XPEXPE
GADD45
GADD45p53R2p53R2
PCNAPCNA
hMSH2hMSH2
p53 family targetp53 family target
p53 target
p53 targetp53 target
p53 targetDamaging Agent
Fig. 7 – DNA repair mediated by p53 family target genes. Some mutations, which are linked to cancer, appear to involve the failure of one or
many of a given cell’s repair systems. One example of such an error involves DNA mismatch repair (MMR). After DNA copies itself, proteins
from mismatch repair genes act as proofreaders to identify and correct mismatches. If a loss or mutation occurs in the mismatch repair genes,
sporadic mutations are more likely to accumulate. Other errors in repair may involve bases or even whole nucleotides being incorrectly cut out
(Nucleotide-excision repair-NER) as repair proteins try to fix DNA after bulky molecules, such as the carcinogens in cigarettes, have attached
themselves. This is classed as faulty excision repair. Any of these mistakes (and others not appearing in the figure shown) may enable mutations topersist, be copied, and eventually contribute to cancer development. Both, MMR and NER, are repair processes mediated by p53 family proteins.
p53 is a transcription factor whose activity is regulated by phosphorylation. The function of p53 is to prevent the cell from progressing through the
cell cycle if DNA damage is found. It may do this in variety of ways; from holding the cell at a checkpoint until repairs can be made, to causing the
cell to enter apoptosis if the damage cannot be repaired. The critical role of p53 is evidenced by the fact that it is mutated in a very large proportion
of tumours from nearly every possible source.
An Acad Bras Cienc (2007) 79(4)

608 PAULA A. OLIVEIRA et al.
carcinogenesis through the application of several chem-
ical, biochemical and biological techniques facilitates
the identification of those basic mechanisms involved
in neoplasic development (King et al. 1995, Maronpot
and Boorman 1996). Experimental assays with labo-ratory animals, epidemiological studies and quick tests
enable the identification of carcinogenic compounds, the
dissection of many aspects of carcinogenesis, and the es-
tablishment of effective strategies to prevent the cancer
which results from exposure to chemicals (Grisham et al.
1984, Butterworth et al. 1992, Maronpot and Boorman1996, Airoldi et al. 1999).
INV I T R O ASSAYS OF CELLTRANSFORMATION
In vitro models are used to study the molecular mecha-nisms inherent to the neoplasic transformation of normalcells (Guengerich 2000, Achanzar et al. 2002). Theseassays use prokaryotic and human cells, have differinglevels of complexity, and can overcome the ethical as-pects related to animal experimentation (Masters 2000).
In 1970, a number of laboratory tests were devel-
oped to evaluate the mutagenic power of different chemi-
cal compounds, with the Ames test gaining particular dis-tinction. This test semi-quantitatively evaluates a chem-ical’s ability to induce mutations in Salmonella tiphy-
murium in a culture medium improved with microso-
matic enzymes (Ames 1984). Between 70 and 90% ofknown chemical carcinogens show positive results on theAmes test. Most mutagenic chemicals in vitro are car-cinogenic in vivo. Due to the high correlation that existsbetween mutagenecity and carcinogenicity, the Ames testis frequently used to evaluate the carcinogenic potentialof chemicals. However, substances such as nitrosaminesand beryllium do not strongly correspond to their resultsin the Ames test (Gonzalez 2001, Payne and Kemp 2003).It has been estimated that at least one hundred methodsof in vitro testing the carcinogenic power of a compoundhave appeared over the last two decades.
Some scientists have questioned whether cells in
culture maintain their bioactivation and detoxificationmechanisms (Masters 2000, Gutiérrez and Salsamendi2001). To validate the results obtained from these as-says it is important to check if these results occur underphysiological conditions considered as normal. To over-come the advantages of these methods, and those pre-viously mentioned regarding in vivo assays, new meth-
ods were developed using human tissues and biologicalfluids to obtain specific biomarkers, which combinedwith the epidemiological studies gave results that aremore reliable. These experiments are labelled as themolecular epidemiology of cancer or molecular dosime-try (Bondy 2004, Yang and Schlueter 2005).
INV I V O ASSAYS OF CARCINOGENESIS
Experimental models with animals have been used suc-cessfully for a number of decades. They have enabledus to understand diseases, to discover etiological fac-tors and to test many treatments (Maronpot and Boor-man 1996). There are innumerable anatomic, physio-
logical and biochemical resemblances between rodents
and humans that justify their use in carcinogenicity test-
ing (Maronpot and Boorman 1996, Balmain and Harris2000). Results obtained from these studies permit theidentification of the harmful carcinogenic compounds inthe absence of real and credible human references andprotect the public health (Huff 1992).
Current strategies to identify the carcinogenic po-
tentiality of certain compounds include experimentalprotocols lasting a minimum of two years (Payne andKemp 2003). These can stretch from 5 to 7 years if wetake into account the posterior analysis of the results ob-tained via the different methods (Tennant et al. 1999).These assay groups of males and females, of mice andrats, are exposed to two or three doses of the agent be-ing tested while a non-exposed (control) group is alsoused (Weisburger 1999). The experiment has a previ-ously established duration and the animals that surviveare sacrificed at the end of the experiment (van Leeuwenand Zonneveld 2001, Pitot 2001, Payne and Kemp 2003).
Animals are examined post-mortem in order to eval-
uate the incidence of neoplasic development and otherpathological changes. Statistical analysis is used to eval-uate if the neoplasic incidence is significantly differentfrom the control group (Ito et al. 1992, Lutz 1998, Ca-margo et al. 1999, Tennant et al. 1999, Payne and Kemp2003). On the cases in which the control animals do notshow neoplasias, the results are considered significant if10% of the animals exposed to the carcinogen developneoplasias (Pitot 2001).
Carcinogenic assays on rodents identify potential
An Acad Bras Cienc (2007) 79(4)

CHEMICAL CARCINOGENESIS 609
carcinogens for humans. Achieving a positive result on
a conventional essay indicates that there exists only a po-
tential danger. Its meaning for human health will depend
on other factors, some of which require additional stud-
ies (Maronpot and Boorman 1996). The extrapolation ofresults obtained via experimental work with rodents is
contested by the following arguments (Gaylor and Chen
1986, Huff 1992, Tennant et al. 1995, Haseman et al.
2001, Waddell 2002):
a) It has not been confirmed if rodent models are rep-
resentative of carcinogenesis in humans.
b) The studies are too long.
c) The doses are too high and may cause a proliferative
response in normal cells.
d) Many of the effects observed in animals have little
importance for man.
e) The protective effects of the organism, metabolic
detoxification, and DNA repair cannot be taken intoaccount once they are overwhelmed by exposure tohigh doses.
f) Synergic effects are not taken into account with
other chemical compounds.
Based on data accumulated from experiments in re-
cent years, and according to Gutiérrez and Salsamendi(2001), they provide the following factors which favourthese assays:
a) All substances that revealed carcinogenic activity in
humans, apart from rare exceptions, are also posi-
tive in rodent assays.
b) Although many chemical carcinogens for animals
do not cause cancer in humans, many of human
carcinogens were discovered from assays in ani-
mals such as: aflotoxins, diethylstilbestrol or vinyl
chloride.
Molecular biology has provided new models with
which to study carcinogenesis with the development of
transgenic and knockout rodents. Some models have
mutations in the ras proto-oncogenes and in the p53-
suppressor gene (Sills et al. 2001, Pitot 2001). Animal
models deficient in p53 protein and ras genes are more
sensitive to the identification of genotoxic carcinogens(Sills et al. 2001). According to Pritchard et al. (2003),
the utilization of transgenic models to identify carcino-
genic compounds has the following advantages:
a) Tumours developed more quickly.
b) The assays are shorter, with a duration of 24 to 26
weeks.
c) Fewer animals are used.
d) Through genetic modification, it is possible to
identify those mechanisms associated with neo-plasic development.
Although these models are promising, they also
have limitations because they can exhibit metabolic al-terations, which are not consistently relevant to carcino-genesis. In addition, mutated genes can influence thenature of neoplasia that is developed, increasing the dif-ficulty of measuring the response in humans (Pritchardet al. 2003).
It is necessary to pay attention to the analysis of
the results, because there is evidence which indicatesthat carcinogens can act through specific mechanisms.The premise that those carcinogenic compounds experi-mentally tested are harmful for man is not always valid(Swenberg et al. 1992, Cohen and Lawson 1995). The
results obtained using rodents act as back-up against any
false negatives obtained through in vitro researches and
can be used to prevent, or reduce, human exposure to asuspected carcinogen (Payne and Kemp 2003).
EPIDEMIOLOGICAL STUDIES
Epidemiological studies provide a great deal of informa-tion about exposure to those chemicals present in food,the environment and at work, but are limited as far as the
identification of etiological factors are concerned, espe-
cially in cases where neoplasic development results fromthe interaction of multiple agents (Garner 1998, Ten-nant 1998, Weinstein 1991). Epidemiological studiesare retrospective and unless a large number of individualsare studied their sensitivity is reduced (Weinstein 1988,Tennant 1998).
Epidemiological techniques have been useful for
identifying exposure to high carcinogenic concentra-
tions. Yet, it is difficult to understand the individual con-tribution of a certain chemical within a complex situa-
An Acad Bras Cienc (2007) 79(4)

610 PAULA A. OLIVEIRA et al.
tion like environmental contamination. Carrying out epi-
demiological studies of a scientific nature is difficult for
several reasons (Farmer 1994, Tennant 1998):
a) The difficulty in evaluating external and internal ex-
position to chemicals.
b) The impossibility of simultaneously controlling ex-
posure to other chemicals, and analysing the influ-
ence of those environmental and physiological fac-
tors that influence the evolution of the disease.
c) The latency period between initial exposure and
cancer development.
Only in some cases, such as with tobacco smoke,
does the epidemiological evidence of cause and effectbe held beyond any doubt (Gutiérrez and Salsamendi2001).
OTHER METHODS
The carcinogenic influence of a substance can be deter-mined using computer programmes that thoroughly sim-ulate man’s physiological and metabolic procedures andrelate them to the molecular configuration of the sub-stance being studied (Loew et al. 1985). These chemicalproperties are related to the molecular structure of chem-ical, physical, and toxicological properties (Barratt andRodford 2001, Feng et al. 2003).
Statistical learning methods have recently been ex-
plored as a new approach for genotoxicity predictionwithout any restrictions on the features of structures ortypes of molecules. Instead of focusing on specific struc-tural features or a particular group of related molecules,these methods classify molecules into genotoxic positiveor non-genotoxic agents based on their general structuraland physicochemical properties, regardless of their struc-tural and chemical types (Li et al. 2005).
Other available tests concern the use of protozoa
cultures and the chorioallantoic membrane. The ciliatedprotozoan Tetrahymena pyriformis may be used in bioas-
says to evaluate the cytotoxic impact of many chemicalcompounds (Bonnet et al. 2003). The chicken chorioal-lantoic membrane assay is used to study angiogenesis
during tumour growth (Tufan and Satiroglu-Tufan 2005).CONCLUSIONS
In summary, our objectives for this article were to re-
view the current information available on chemical carci-
nogenesis. Chemical carcinogenesis is a multistage and
multicausal process in which normal cells become firstinitiated, then malignant and invasive. Each of thesestages is exceedingly complex in itself. The acquisition
of the capacity to survive and grow independently from
other cells represents a crucial event in the mechanism
of cancer development. Most of the morphological, bio-chemical and genetic changes currently observed shouldbe considered as the expression of the adaptation of neo-plasic cells to survive in a familiar but hostile environ-ment. The prediction of chemical carcinogenicity is ofgreat importance to human risk assessment.
ACKNOWLEDGMENTS
Grant support for this study was provide by Fundação
para a Ciência e Tecnologia, Ministério da Ciência eEnsino Superior, Portugal (number 12453/2003).
RESUMO
A sociedade obtém numerosos benefícios da utilização de
compostos químicos. A aplicação dos pesticidas, por exem-
plo, permitiu obter alimento em quantidade suficiente para
satisfazer as necessidades alimentares de milhões de pessoas,condição relacionada com o aumento da esperança de vida.Os benefícios estão, por vezes associados a desvantagens, os
efeitos resultantes da exposição a compostos químicos enqua-
dram-se entre a morte imediata e um longo processo de car-cinogênese química. A carcinogênese química inclui três eta-
pas definidas como iniciação, promoção e progressão. Cada
uma delas caracteriza-se por transformações morfológicas ebioquímicas, e resulta de alterações genéticas e/ou epigenéti-cas. No grupo das alterações genéticas incluem-se mutações
nos genes que controlam a proliferação celular, a morte celular
e a reparação do DNA – i.e. mutações nos proto-oncogenes egenes supressores de tumor. Os fatores epigenéticos, também
considerados como caracteres não genéticos, podem contribuir
para a carcinogênese por mecanismos de silenciamento gênico.A utilização de diferentes metodologias possibilita o reconhe-cimento e a compreensão dos mecanismos básicos envolvidos
no desenvolvimento do cancro. Ensaios experimentais com
animais de laboratório, estudos epidemiológicos e alguns testes
An Acad Bras Cienc (2007)
79(4)

CHEMICAL CARCINOGENESIS 611
rápidos permitem identificar compostos carcinogênicos, ana-
lisar os eventos envolvidos na carcinogênese e estabelecer es-
tratégias para prevenir a exposição a estes agentes.
Palavras-chave: etapas da carcinogênese, avaliação de carci-
nogeneicidade, carcinogênicos químicos, carcinogênese quí-
mica.
REFERENCES
ACHANZAR WE, B RAMBILA EM, D IWAN BA, W EBBER
MM AND WAALKES MP. 2002. Inorganic arsenite-
induced malignant transformation of human prostate epi-thelial cells. J Natl Cancer Inst 94: 1888–1891.
A
DIMOOLAM SAND FORD JM. 2002. p53 and DNA dam-
ageinducible expression of the xeroderma pigmentosumgroup C gene. Proc Natl Acad Sci USA 99: 12985–12990.
A
IROLDI L, P ASTORELLI R, M AGAGNOTTI CAND FANELLI
R. 1999. Carcinogen-DNA adducts as tools in risk assess-
ment. Adv Exp Med Biol 472: 231–240.
AMES BN. 1984 The detection of environmental mutagens
and potential carcinogens. Cancer 53: 2034–2040.
ASHBY J. 1996. Prediction of Rodent Carcinogenicity for 30
Chemicals. Environ Health Perspect 104S: 1101–1104.
BABENKO VN, B ASU MK, K ONDRASHOV FA, R OGOZIN
IBAND KOONIN EV. 2006. Signs of positive selection
of somatic mutations in human cancers detected by ESTsequence analysis. BMC Cancer 9: 26–36.
B
AIRD WM AND MAHADEV AN B. 2004. The uses of carci-
nogen-DNA adduct measurement in establishing mecha-nisms of mutagenesis and in chemoprevention. Mutat Res547: 1–4.
B
ALMAIN AAND HARRIS CC. 2000. Carcinogenesis in
mouse and human cells: parallels and paradoxes. Carci-nogenesis 21: 371–377.
B
ARRATT MD AND RODFORD RA. 2001. The computational
prediction of toxicity. Opin Chem Biol 5: 383–388.
BARRETT JC. 1993. Mechanisms of multistep carcinogene-
sis and carcinogen risk assessment. Environ Health Per-spect 100: 9–20.
B
ARRETT JCAND ANDERSON M. 1993. Molecular mecha-
nisms of carcinogenesis in humans and rodents. MolCarcinog 7: 1–13.
B
ARRET JCAND WISEMAN RW. 1987. Cellular and molec-
ular mechanisms of multistep carcinogenesis: relevanceto carcinogen risk assessment. Environ Health Perspect76: 65–70.
B
ARTSCH HAND HIETANEN E. 1996. The role of individualsusceptibility in cancer burden related to environmental
exposure. Environ Health Perspect 104: 569–577.
BEREMBLUM IAND SHUBIK P. 1947. The role of croton oil
applications, associated with a single painting of a carcino-gen, in tumor induction of the mouse’s skin. Br J Cancer1: 379–382.
B
ERTRAM JS. 2001. The molecular biology of cancer. Mol
Aspects Med 21: 167–223.
BLAGOSKLONNY MV. 2005. Molecular theory of cancer.
Cancer Biol Ther 4: 621–627.
BOLT HM, F OTH H, H ENGSTLER JG AND DEGEN GH.
2004. Carcinogenicity categorization of chemicals-newaspects to be considered in a European perspective. Toxi-col Lett 151: 29–41.
B
ONDY M. 2004. Estimated risk in malignancy: the emerging
field of molecular epidemiology. Clin Adv Hematol Oncol2: 147–151.
B
ONNET JL, D USSER M, B OHATIER JAND LAFFOSSE J.
2003. Cytotoxic assessment of three therapeutic agents,cyclosporine-A, cisplatin and doxorubicin, with the cili-ated protozoan Tetrahymena pyriformis. Res Microbiol154: 375–385.
B
UTTERWORTH BE AND BOGDANFFY MS. 1999. A com-
prehensive approach for integration of toxicity and cancerrisk assessments. Regul Toxicol Pharmacol 29: 23–36.
B
UTTERWORTH BE, P OPPJA, C ONOLLY RB AND GOLDS –
WORTHY TL. 1992. Chemically induced cell prolifera-
tion in carcinogenesis. IARC Sci Publ 116: 279–305.
BUTTERWORTH BE, T EMPLIN MV, C ONSTAN AA, S PRAN –
KLE CS, W ONG BA, P LUTA LJ, E VERITT JIAND
RECIO L. 1998. Long-term mutagenicity studies with
chloroform and dimethylnitrosamine in female lacI trans-genic B6C3F1 mice. Environ Mol Mutagen 31: 248–56.
C
AMARGO JLV, S ALV ADORI DMF, R OCHA NS, B AEBISAN
LF AND RIBEIRO LR. 1999. The detection of chemical
carcinogens in an alternative medium-term bioassay. JBraz Ass Advan Science 51: 22–26.
C
ARRIER FET AL . 1999. Gadd45, a p53-responsive stress
protein, modifies DNA accessibility on damaged chro-matin. Mol Cell Biol 19: 1673–1685.
C
HAO EC AND LIPKIN SM. 2006. Molecular models for
the tissue specificity of DNA mismatch repair-deficientcarcinogenesis. Nucleic Acids Research 34: 840–852.
C
OHEN SM. 1991. Analysis of modifying factors in chemical
carcinogenesis. Prog Exp Tumor Res 33: 21–40.
COHEN SM. 1995. Role of urinary physiology and chemistry
in bladder carcinogenesis. Food Chem Toxicol 33: 715–730.
An Acad Bras Cienc (2007)
79(4)

612 PAULA A. OLIVEIRA et al.
COHEN SM. 1998. Cell proliferation and carcinogenesis.
Drug Metab Rev 30: 339–357.
COHEN SM AND ELLWEIN LB. 1991. Genetic errors, cell
proliferation, and carcinogenesis. Cancer Res 51: 6493–6505.
C
OHEN SM AND LAWSON TA. 1995. Rodent bladder tumors
do not always predict for humans. Cancer Lett 93: 9–16.
COHEN SM, P URTILO DT AND ELLWEIN LB. 1991. Ideas
in pathology. Pivotal role of increased cell proliferationin human carcinogenesis. Mod Pathol 4: 371–382.
C
OHEN SM, G ARLAND EM AND ELLWEIN LB. 1992. Can-
cer enhancement by cell proliferation. Prog Clin BiolRes 374: 213–229.
C
ONNOLY RB, R EITZ RH, C LEWELL 3RDHJAND ANDER –
SON ME. 1988. Pharmacokinetics, biochemical mecha-
nism and mutation accumulation: a comprehensive modelof chemical carcinogenesis. Toxicol Lett 43: 189–200.
C
OSTA M, Y ANY, Z HAO DAND SALNIKOW K. 2003.
Molecular mechanisms of nickel carcinogenesis: gene si-lencing by nickel delivery to the nucleus and gene acti-vation/inactivation by nickel-induced cells signalling. JEnviron Monit 5: 222–223.
D
EWHIRST MW, L ORA-MICHIELS M, V IGLIANTI BL,
DEWEY WC AND REPACHOLI M. 2003. Carcinogenic
effects of hyperthermia. Int J Hyperthermia 19: 236–251.
DIXON KAND KOPRAS E. 2004. Genetic alterations and
DNA repair in human carcinogenesis. Semin Cancer Biol14: 441–448.
D
RABLOS FET AL . 1998. Studies of initiation and promo-
tion of carcinogenesis by N-nitroso compounds. CancerLett 123: 185–191.
D
YBDAHL M, F RENTZ G, V OGEL U, W ALLIN HAND NEXO
BA. 1999. Low DNA repair is a risk factor in skin carci-
nogenesis: a study of basal cell carcinoma in psoriasispatients. Mutat Res 433: 15–22.
D
YBING EAND SANNER T. 1999. Species differences in
chemical carcinogenesis of the thyroid gland, kidney andurinary bladder. IARC Sci Publ 147: 15–32.
F
ARBER E. 1984. The multi-step nature of cancer develop-
ment. Cancer Res 44: 4217–4223.
FARMER PB. 1994. Carcinogen adducts: use in diagnosis and
risk assessment. Clin Chem 40: 1438–1443.
FENG J, L URATI L, O UYANG H, R OBINSON T, W ANG Y,
YUAN SAND YOUNG SS. 2003. Predictive toxicology:
benchmarking molecular descriptors and statistical meth-ods. J Chem Inf Comput Sci 43: 1463–1470.F
LORES -ROZAS H, C LARK DAND KOLODNER RD. 2000.
Proliferating cell nuclear antigen and Msh2p-Msh6p in-teract to form an active mispair recognition complex. NatGenet 26: 375–378.
F
OULDS L. 1954. The experimental study of tumor progres-
sion: a review. Cancer Res 14: 327–339.
FRIEDBERG EC. 2003. DNA damage and repair. Nature 421:
436–440.
FROWEIN J. 2000. Hypothesis: chemical carcinogenesis me-
diated by a transiently active carcinogen receptor. Cyto-genet Cell Genet 91: 102–104.
G
ALATI G, T ENG S, M ORIDANI MY, C HAN TS AND
O’B RIEN PJ. 2000. Cancer chemoprevention and apop-
tosis mechanisms induced by dietary polyphenolics. DrugMetabol Drug Interact 17: 311–349.
G
ARCEA G, D ENNISON AR, S TEWARD WP AND BERRY
DP. 2003. Chemoprevention of gastrointestinal malig-
nancies. ANZ J Surg 73: 680–686.
GARNER RC. 1998. The role of DNA adducts in chemical
carcinogenesis. Mutat Res 402: 67–75.
GAYLOR DW AND CHEN JJ. 1986. Relative potency of chem-
ical carcinogens in rodents. Risk Anal 6: 283–290.
GOLKA K, K OPPS SAND MYSLAK ZW. 2004. Carcino-
genicity of azo colorants: influence of solubility and bio-availability. Toxicol Lett 151: 203–210.
G
OMES -CARNEIRO MR, R IBEIRO -PINTO LF AND PAUM –
GARTTEN FJ. 1997. Environmental risk factors for gastric
cancer: the toxicologist’s standpoint. Cad Saúde Pública13 (Suppl): 27–38.
G
ONZALEZ FJ. 2001. The use of gene knockout mice to
unravel the mechanisms of toxicity and chemical carcino-genesis. Toxicol Lett 120: 199–208.
G
ONZALEZ FJAND KIMURA S. 2001. Understanding the
role of xenobiotic-metabolism in chemical carcinogenesisusing gene knockout mice. Mutat Res 477: 79–87.
G
RISHAM JW, K AUFMANN WK AND KAUFMAN DG. 1984.
The cell cycle and chemical carcinogenesis. Surv SynthPatho Res 1: 49–66.
G
UENGERICH FP. 2000. Metabolism of chemical carcino-
gens. Carcinogenesis 21: 345–351.
GUENGERICH FP. 2001. Forging the links between metabol-
ism and carcinogenesis. Mutat Res 488: 195–209.
GUITTET O, H AKANSSON P, V OEVODSKAYA N, F RIDD
S, G RASLUND A, A RAKAWA H, N AKAMURA YAND
THELANDER L. 2001. Mammalian p53R2 protein forms
an active ribonucleotide reductase in vitro with the R1
An Acad Bras Cienc (2007) 79(4)

CHEMICAL CARCINOGENESIS 613
protein, which is expressed both in resting cells
in response to DNA damage and in proliferating cells.J Biol Chem 276: 40647–40651.
G
UTIÉRREZ JBAND SALSAMENDI AL. 2001. Fundamien-
tos de ciência toxicológica. Diaz de Santos, Madrid, p.155–177.
H
ANAHAN DAND WEINBERG RA. 2000. The hallmarks of
cancer. Cell 100: 57–70.
HANAWALT PC, F ORD JM AND LLOYD DR. 2003. Func-
tional characterization of global genomic DNA repair andits implications for cancer. Mutat Res 544: 107–114.
H
ARMS K, N OZELL SAND CHEN X. 2004. The common
and distinct target genes of the p53 family transcriptionfactors. Cell Mol Life Sci 61: 822–842.
H
ARTWIG A, A SMUSS M, E HLEBEN I, H ERZER U, K OSTE –
LAC D, P ELZER A, S CHWERDTLE TAND BURKLE A.
2002. Interference by toxic metal ions with DNA repairprocesses and cell cycle control: molecular mechanisms.Environ Health Perspect 110 (Suppl 5): 797–799.
H
ASEGAWA R, F UTAKUCHI M, M IZOGUCHI Y, Y AMA –
GUCHI T, S HIRAI T, I TONAND LIJINSKY W. 1998.
Studies of initiation and promotion of carcinogenesis byN-nitroso compounds. Cancer Lett 123: 185–191.
H
ASEMAN J, M ELNICK R, T OMATIS LAND HUFFJ. 2001.
Carcinogenesis bioassays: study duration and biologicalrelevance. Food Chem Toxicol 39: 739–744.
H
AWIGHORST T, V ELASCO P, S TREIT M, H ONG YK,
KYRIAKIDES TR, B ROWN LF, B ORNSTEIN PAND
DETMAR M. 2001. Thrombospondin-2 plays a protective
role in multistep carcinogenesis: a novel host anti-tumordefense mechanism. EMBO J 20: 2631–2640.
H
AYES RB. 1995. Genetic susceptibility and occupational
cancer. Med Lav 86: 206–213.
HAYES JD AND PULFORD DJ. 1995. The glutathione S-
transferase supergene family: regulation of GST and thecontribution of the isoenzymes to cancer chemoprotectionand drug resistance. Crit Rev Biochem Mol Biol 30: 445–600.
H
EIDELBERGER C. 1977. Chemical carcinogenesis. Cancer
40: 430–433.
HENGSTLER JG, A RAND M, H ERRERO ME AND OESCH
F. 1998. Polymorphisms of N-acetyltransferases, gluta-
thione S-transferases, microsomal epoxide hydrolase andsulfotransferases: influence on cancer susceptibility. Re-cent Results Cancer Res 154: 47–85.
H
UFFJ. 1992. Chemical toxicity and chemical carcinogenesis.
Is there a causal connection? A comparative morpholog-ical evaluation of 1500 experiments. IARC Sci Pub 116:
437–475.
HUFFJ. 1994. Chemicals causally associated with cancers in
humans and in laboratory animals. A perfect concordance.In: W
ALLACE MP AND WARD JM (Eds), Carcinogene-
sis, Raven Press, Ltd., New York, p. 25–37.
HUFF J. 1999. Chemicals associated with tumours of the
kidney, urinary bladder and thyroid gland in laboratoryrodents from 2000 US National Toxicology Program /National Cancer Institute bioassays for carcinogenicity.IARC Sci Pub 147: 211–225.
H
WANG BJ, F ORD JM, H ANAWALT PC AND CHUG. 1999.
Expression of the p48 xeroderma pigmentosum gene isp53-dependent and is involved in global genomic repair.Proc Natl Acad Sci USA 96: 424–428.
I
SHIKAWA T, I DEF, Q INX, Z HANG S, T AKAHASHI Y,
SEKIGUCHI M, T ANAKA KAND NAKATSURU Y. 2001.
Importance of DNA repair in carcinogenesis: evidencefrom transgenic and gene targeting studies. Mutat Res477: 414–419.
I
TON, S HIRAI TAND HASEGAWA R. 1992. Medium-term
bioassays for carcinogens. IARC Sci Publ 116: 353–388.
JENG JH, C HANG MC AND HAHN LJ. 2001. Role of areca
nut in betel quid-associated chemical carcinogenesis: cur-rent awareness and future perspectives. Oral Oncol 37:477–492.
K
HAN QA AND DIPPLE A. 2000. Diverse chemical carcino-
gens fail to induce G(1) arrest in MCF-7 cells. Carcino-genesis 21: 1611–1618.
K
HAN QA, V OUSDEN KH AND DIPPLE A. 1999. Lack of
p53-mediated G1 arrest in response to an environmentalcarcinogen. Oncology 57: 258–264.
K
INGC, W ANG C, G ORELICK NAND FREDERICK C. 1995.
Genotoxicity in the rodent urinary bladder. Food ChemToxicol 33: 757–769.
K
INZLER KW AND VOGELSTEIN B. 1997. Cancer-suscept-
ibility genes. Gatekeepers and caretakers. Nature 386:761–763.
K
LAUNIG JE, K AMENDULIS LM AND XUY. 2000. Epige-
netic mechanisms of chemical carcinogenesis. Hum ExpToxicol 19: 543–555.
K
OIVUSALO M, J AAKKOLA JJ, V ARTIAINEN T, H AKULI –
NEN T, K ARJALAINEN S, P UKKALA EAND TUOMISTO
J. 1994. Drinking water mutagenicity and gastrointestinal
and urinary tract cancers: an ecological study in Finland.Am J Public Health 84: 1223–1228.
L
AICAND SHIELDS PG. 1999. The role of interindividual
variation in human carcinogenesis. J Nutr 29: 552S–555S.
An Acad Bras Cienc (2007) 79(4)

614 PAULA A. OLIVEIRA et al.
LAMERS MH, P ERRAKIS A, E NZLIN JH, W INTERWERP
HH, W INDNAND SIXMA TK. 2000. The crystal struc-
ture of DNA mismatch repair protein MutS binding to aG×T mismatch. Nature 407: 711–717.
L
IH, U NGCY, Y APCW, X UEY, L IZR, C AOZW AND
CHEN YZ. 2005. Prediction of genotoxicity of chemical
compounds by statistical learning methods. Chem ResToxicol 18: 1071–1080.
L
OCK EA, R EED CJ, M CMILLAN JM, O ATIS JE J R AND
SCHNELLMANN RG. 2007. Lack of formic acid produc-
tion in rat hepatocytes and human renal proximal tubulecells exposed to chloral hydrate or trichloroacetic acid.Toxicology 230: 234–243.
L
OEB LA. 1998. Cancer cells exhibit a mutator phenotype.
Adv Cancer Res 72: 25–56.
LOEW GH, P OULSEN M, K IRKJIAN E, F ERRELL J,
SUDHINDRA BSAND REBAGLIATI M. 1985. Computer-
assisted mechanistic structure-activity studies: applica-tion to diverse classes of chemical carcinogens. EnvironHealth Perspect 61: 69–96.
L
UCH A. 2005. Nature and nurture – lessons from chemical
carcinogenesis. Nat Rev Cancer 5: 113–125.
LUTZ WK. 1998. Dose-response relationships in chemical
carcinogenesis: superposition of different mechanismsof action, resulting in linear-nonlinear curves, practicalthresholds, J-shapes. Mutat Res 405: 117–124.
L
UTZ WK. 1999. Dose-response relationships in chemical
carcinogenesis reflect differences in individual suscepti-bility. Consequences for cancer risk assessment, extrapo-lation, and prevention. Hum Exp Toxicol 18: 707–712.
L
UTZ WK. 2000. A true threshold dose in chemical carcino-
genesis cannot be defined for a population, irrespective ofthe mode of action. Hum Exp Toxicol 19: 566–568.
L
UTZWK. 2001. Susceptibility differences in chemical carci-
nogenesis linearize the dose-response relationship: thresh-old doses can be defined only for individuals. Mutat Res482: 71–76.
L
UTZ WK. 2002. Differences in individual susceptibility to
toxic effects of chemicals determine the dose-responserelationship and consequences of setting exposure stan-dards. Toxicol Lett 126: 155–158.
M
ARONPOT RR. 1996. A symposium summary and perspec-
tive on comparative molecular biology of cancer. ToxicolPathol 24: 801–814.
M
ARONPOT RR AND BOORMAN GA. 1996. The contribu-
tion of the mouse in hazard identification studies. ToxicolPathol 24: 726–731.M
ASTERS JR. 2000. Human cancer cell lines: fact and fan-
tasy. Nat Rev Mol Cell Biol 1: 233–236.
MEHTA R. 1995. The potential for the use of cell proliferation
and oncogene expression as intermediate markers duringliver carcinogenesis. Cancer Lett 93: 85–102.
M
ELNICK RL, H UFF J, B ARRETT JC, M ARONPOT RR,
LUCIER GAND PORTIER CJ. 1993. Cell proliferation
and chemical carcinogenesis: symposium overview. En-viron Health Perspect 101: 3–7.
M
ELNICK RL, K OHN MC AND PORTIER CJ. 1996. Impli-
cations for risk assessment of suggested non-genotoxicmechanisms of chemical carcinogenesis. Environ HealthPerspect 104: 123–134.
M
ILLER EC AND MILLER JA. 1952. In vivo combinations
between carcinogens and tissue constituints and theirpossible role in carcinogenesis. Cancer Res 12: 547–556.
M
ILLER JAAND MILLER EC. 1975. Metabolic activation
and reactivity of chemical carcinogens. Mutat Res 33:25–26.
M
ILLER 3RDMC, M OHRENWEISER HW AND BELL DA.
2001. Genetic variability in susceptibility and response totoxicants. Toxicol Lett 120: 269–280.
M
INAMOTO T, M AIMAND RONAI Z. 2000. K-ras mutation:
early detection in molecular diagnosis and risk assessmentof colorectal, pancreas, and lung cancers-a review. CancerDetect Prev 24: 1–12.
M
IRSALIS JC, S TEINMETZ KL, H AMILTON CM, B AKKE
JPAND GARIN KE. 1990. The role of cell proliferation
in chemical carcinogenesis. Prog Clin Biol Res 340: 113–122.
M
OSTAFA MH, S HEWEITA SA AND O’C ONNOR PJ. 1999.
Relationship between schistosomiasis and bladder cancer.Clin Microbiol Rev 12: 97–111.
M
OUSTACCHI E. 1998. [Molecular mechanisms of carcino-
genesis: the role of systems of DNA repair]. Bull AcadNatl Med 182: 33–46.
N
AKANO K, B ALINT E, A SHCROFT MAND VOUSDEN KH.
2000. A ribonucleotide reductase gene is a transcriptionaltarget of p53 and p73. Oncogene 19: 4283–4289.
N
GUYEN -BAGAND VASSEUR P. 1999. Epigenetic events
during the process of cell transformation induced by car-cinogens (review). Oncol Rep 6: 925–932.
O
DAY. 2004. Analysis of the involvement of human N-acetyl-
transferase 1 in the genotoxic activation of bladder carci-nogenic arylamines using a SOS/umu assay system. MutatRes 554: 399–406.
An Acad Bras Cienc (2007)
79(4)

CHEMICAL CARCINOGENESIS 615
OESCH F, H ERRERO ME, H ENGSTLER JG, L OHMANN M
AND ARAND M. 2000. Metabolic detoxification: impli-
cations for thresholds. Toxicol Pathol 28: 382–387.
OHSHIMA H, T ATEMICHI MAND SAWA T. 2003. Chemi-
cal basis of inflammation-induced carcinogenesis. ArchBiochem Biophys 417: 3–11.
O
HSHIMA H, T AZAWA H, S YLLA BS AND SAWA T. 2005.
Prevention of human cancer by modulation of chronicinflammatory processes. Mutat Res 591: 110–122.
O
TTENEDER MAND LUTZ WK. 1999. Correlation of DNA
adduct levels with tumor incidence: carcinogenic potencyof DNA adducts. Mutat Res 424: 237–247.
P
ARK BK, K ITTERINGHAM NR, M AGGS JL, P IRMOHA –
MED MAND WILLIAMS DP. 2005. The role of meta-
bolic activation in drug-induced hepatotoxicity. Annu RevPharmacol Toxicol 45: 177–202.
P
AYNE SR AND KEMP CJ. 2003. p27(Kip1) (Cdkn1b)-defi-
cient mice are susceptible to chemical carcinogenesis andmay be a useful model for carcinogen screening. ToxicolPathol 31: 355–363.
P
ITOT HC. 2001. Animal models of neoplastic development.
Dev Biol (Basel) 106: 53–57.
PITOT HC AND DRAGAN YP. 1991. Facts and theories con-
cerning the mechanisms of carcinogenesis. FASEB J 5:2280–2286.
P
LAYER A, B ARRETT JCAND KAWASAKI ES. 2004. Laser
capture microdissection, microarrays and the precise defi-nition of a cancer cell. Expert Rev Mol Diagn 4: 831–840.
P
OIRIER MC, S ANTELLA RM AND WESTON A. 2000. Car-
cinogen macromolecular adducts and their measurement.Carcinogenesis 21: 353–359.
P
OTTER VR. 1978. Phenotypic diversity in experimental hep-
atomas: the concept of partially blocked ontogeny. The10
thWalter Hubert Lecture. Br J Cancer 38: 1–23.
PRITCHARD JB, F RENCH JE, D AV I S BJ AND HASEMAN
JK. 2003. The role of transgenic mouse models in car-
cinogen identification. Environ Health Perspect 111: 444–454.
Q
UW, B ORTNER CD, S AKURAI T, H OBSON MJ AND
WAALKES MP. 2002. Acquisition of apoptotic resis-
tance in arsenic-induced malignant transformation: roleof the JNK signal transduction pathway. Carcinogenesis
23: 151–159.
R
ICHARDSON FC, B OUCHERON JA, D YROFF MC, P OPP
JAAND SWENBERG JA. 1986. Biochemical and mor-
phologic studies of heterogeneous lobe responses in hep-
atocarcinogenesis. Carcinogenesis 7: 247–251.ROBBINS DAND COTRAN R. 2005. Pathologic basis of dis-
ease. 7thed., Philadelphia: Elsevier Saunders, p. 319–323.
ROJAS M, C ASCORBI I, A LEXANDROV K, K RIEK E, A U-
BURTIN G, M AYER L, K OPP-SCHNEIDER A, R OOTS I
AND BARTSCH H. 2000. Modulation of benzo[a]pyrene
diolepoxide-DNA adduct levels in human white bloodcells by CYP1A1, GSTM1 and GSTT1 polymorphism.Carcinogenesis 21: 35–41.
S
ANTELLA RM, G AMMON M, T ERRY M, S ENIE R, S HEN J,
KENNEDY D, A GRAWAL M, F ARAGLIA BAND ZHANG
F. 2005. DNA adducts, DNA repair genotype/phenotype
and cancer risk. Mutat Res 592: 29–35.
SARASIN AAND MEUNIER -ROTIVAL M. 1976. How chem-
icals may induce cancer. Biomedicine 24: 306–316.
SCHERER SJ, M AIER SM, S EIFERT M, H ANSELMANN
RG, Z ANG KD, M ULLER -HERMELINK HK, A NGEL P,
WELTER C, S CHARTL M. 2000. p53 and c- Jun function-
ally synergize in the regulation of the DNA repair genehMSH2 in response to UV . J Biol Chem 275: 37469–37473.
S
CHMAHL D. 1976. Combination effects in chemical carcino-
genesis (experimental results). Oncology 33: 73–76.
SCHUT HA AND CASTONGUAY A. 1984. Metabolism of car-
cinogenic amino derivatives in various species and DNAalkylation by their metabolites. Drug Metab Rev 15: 753–839.
S
COTT RE, W ILLE JRJJAND WIERML. 1984. Mecha-
nisms for the initiation and promotion of carcinogenesis:a review and a new concept. Mayo Clin Proc 59: 107–117.
S
HACTER EAND WEITZMAN SA. 2002. Chronic inflamma-
tion and cancer. Oncology 6: 217–226.
SHARMA RA AND FARMER PB. 2004. Biological relevance
of adduct detection of the chemopreventation of cancer.Clin Cancer Res 10: 4901–4912.
S
HIH, H UDSON LG AND LIUKJ. 2004. Oxidative stress
and apoptosis in metal ion-induced carcinogenesis. FreeRadic Biol Med 37: 582–593.
S
ILLS RC, F RENCH JEAND CUNNINGHAM ML. 2001. New
models for assessing carcinogenesis: an ongoing process.Toxicol Lett 120: 187–198.
S
IMONS JW. 1995. Genetic, epigenetic, dysgenetic, and non-
genetic mechanisms in tumorigenesis. Crit Rev Oncog 6:261–273.
S
IMONS JW. 1999. Genetic, epigenetic, dysgenetic and non-
genetic mechanisms in tumorigenesis. II. Further delin-eation of the rate limiting step. Anticancer Res 19: 4781–4789.
An Acad Bras Cienc (2007)
79(4)

616 PAULA A. OLIVEIRA et al.
SOBELS FH. 1975. Charlotte Auerbach and chemical muta-
genesis. Mutation Res 29: 171–180.
STENBÄCK F, P ETO RAND SHUBIK P. 1981. Initiation and
promotion at different ages and doses in 2200 mice. I.Methods, and the apparent persistence of initiated cells.Br J Cancer 44: 1–14.
S
TRAUB KM AND BURLINGAME AL. 1981. Carcinogen
binding to DNA. Biomed Mass Spectrom 8: 431–435.
SWENBERG JA, F EDTKE N, C IROUSSEL F, B ARBIN AAND
BARTSCH H. 1992. Etheno adducts formed in DNA of
vinyl chloride-exposed rats are highly persistent in liver.Carcinogenesis 13: 727–729.
T
ANTAND CHUG. 2002. p53 Binds and activates the xero-
derma pigmentosum DDB2 gene in humans but not mice.Mol Cell Biol 22: 3247–3254.
T
ENNANT RW. 1998. Evaluation and validation issues in the
development of transgenic mouse carcinogenicity bioas-says. Environ Health Perspect 106: 473–476.
T
ENNANT RW, F RENCH JEAND SPALDING JW. 1995. Iden-
tifying chemical carcinogens and assessing potential riskin short-term bioassays using transgenic mouse models.Environ Health Perspect 103: 942–950.
T
ENNANT RW, S TASIEWICZ S, M ENNEAR J, F RENCH JE
AND SPALDING JW. 1999. Genetically altered mouse
models for identifying carcinogens. IARC Sci Publ 146:123–150.
T
OTH B. 2001. Species susceptibilities to chemical carcino-
genes: a critical appraisal of the roles of genetic and viralagents. In Vivo 15: 467–478.
T
ROSKO JE. 2001. Commentary: is the concept of “tumor
promotion” a useful paradigm? Mol Carcinog 30: 131–137.
T
ROSKO JE. 2003. The role of stem cells and gap junctional
intercellular communication in carcinogenesis. J BiochemMol Biol 36: 43–48.
T
UFAN AC AND SATIROGLU -TUFAN NL. 2005. The chick
embryo chorioallantoic membrane as a model system forthe study of tumor angiogenesis, invasion and develop-ment of anti-angiogenic agents. Curr Cancer Drug Targets5: 249–266.
VA N LEEUWEN IM AND ZONNEVELD C. 2001. From expo-
sure to effect: a comparison of modeling approaches tochemical carcinogenesis. Mutat Res 489: 17–45.V
OGELSTEIN BAND KINZLER KW. 2004. Cancer genes and
the pathways they control. Nat Med 10: 789–799.
WADDELL WJ. 2002. Thresholds of carcinogenicity of fla-
vors. Toxicol Sci 68: 275–279.
WANG TC, C HIOU CM AND CHANG YL. 1998. Genetic
toxicity of N-methylcarbamate insecticides and their N-nitroso derivatives. Mutagenesis 13: 405–408.
W
EINSTEIN IB. 1988. Strategies for inhibiting multistage car-
cinogenesis based on signal transduction pathways. MutatRes 202: 413–420.
W
EINSTEIN IB. 1991. Cancer prevention: recent progress
and future opportunities Cancer Res 51: 5080–5085.
WEISBURGER JH. 1998. Worldwide prevention of cancer
and other chronic diseases based on knowledge of mech-anisms. Mutat Res 402: 331–337.
W
EISBURGER JH. 1999. Carcinogenicity and mutagenicity
testing, then and now. Mutat Res 437: 105–112.
WILDCP, G ARNER RC, M ONTESANO RAND TURSI F.
1986. Aflatoxin B1 binding to plasma albumin and liverDNA upon chronic administration to rats. Carcinogenesis7: 853–858.
W
ILLIAMS GM. 2001. Mechanisms of chemical carcino-
genesis and application to human cancer risk assessment.Toxicology 161: 3–10.
X
UJAND MORRIS GF. 1999. p53-mediated regulation of
proliferating cell nuclear antigen expression in cells ex-posed to ionizing radiation. Mol Cell Biol 19: 12–20.
Y
AMAGIWA KAND ICHIKAWA K. 1918. Experimental study
of the pathogenesis of carcinoma. J Cancer Res 3: 1–29.
YANG MAND SCHLUETER R. 2005. Molecular Epidemiol-
ogy of Lung Cancer in Female Passive Smokers. J EnvironSci Health C Environ Carcinog Ecotoxicol Rev 23: 75–97.
Y
USPA SH AND POIRIER MC. 1988. Chemical carcino-
genesis: from animal models to molecular models in onedecade. Adv Cancer Res 50: 25–70.
Y
USPA SH, H ENNINGS H, L ICHTI UAND KULESZ -MAR-
TINM. 1983. Organ specificity and tumor promotion.
Basic Life Sci 24: 157–171.
An Acad Bras Cienc (2007) 79(4)