Raymond W. Tennant
Although there is compelling evidencefor multiple mutageniceventsinthe inductionofcancers,there is also
substantial evidence in support of nonmutagenic mechanisms. It is proposed that the genetic basis of noninducedor spontaneous tumors, aswellascancersinduced bynonmutagens,involves heritable changes in theregulation ofgeneexpression.
Cancer isadisease ofenvironment andgenetics. There
is a strong scientific consensus, codified by the
Interna-tional Agencyfor Research on Cancer
environmentalfactors such assmoking, exposure to
sun-light, exposure to certain organic chemicals, and other
occupational and environmental factors establish a solid
basis foranenvironmental component in theinduction of cancers. Data derived from epidemiological studies have
identified differences in the prevalence andtypes of
in the rate or frequency of the development of specific cancers in migrant populations provides support for an
Likewise, there is compelling evidence for a genetic
basis ofhumancancer.Thisincludes evidence forheritable
susceptibilities between populations ofhumans, for
exam-ple, skin cancer among fair skinned Anglo-Saxons and
evidenceforgenetic mechanisms ininduced carcinogene-sis. Theroleofgeneticmechanisms incarcinogenesiswas
firstproposed by Theodore Boveri in 1924 (4), who articu-lated the earliest version of the somatic mutation hypoth-esis.Subsequently,otherevidence for the role ofmutagenic changes in carcinogenesis have come from a variety of
sources.These lines of evidence include the chromosomal
alterations thathave been identifled in manyrodent and
human tumors. Infact, most human and rodent tumors that have been examined showgeneralized chromosomal damage as well as specific chromosomal mutations or
translocations(5). Also,overthepastdecade, evidencehas
emerged associating up to ahundred different dominant
genes (i.e., oncogenes) with carcinogenesis. The role of mutations in the activation ofthese geneshasprovideda
National Institute of EnvironmentalHealth Sciences, Research
Tri-angle Park,NC 27709.
genetic basis for bothinitiating andpromoting events in tumor development (6). Subsequently, evidence has emerged for anti-oncogenes or tumor-suppressor genes, which have an importantregulatoryrole incontrollingthe
expression or function of oncogenes (7). The function of these suppressor genes can be lost through mutation or translocation. Subsequent observations have provided a
schemeinwhichmultiple genetic changescanbe identified
and associated with sequential alterations giving rise to tumorsinhumans(8).
These observations represent a very brief and only a
partiallist of the data that areavailableto support arole forspecificgenes incarcinogenesisand forspecific altera-tions ormutations inthose genes that initiate or promote the carcinogenic process. Inthe face of such compelling data,it isdifficult to consider thepossibilitythatthere may be alternativemechanisms ofcarcinogenesis.
One of the strongest lines of evidence in support of nonmutagenicmechanisms arethechemicals that demon-strate no consistent mutagenic properties, yet have the
capacity to induce tumors in rodent bioassays (9). Our
operational definition of a nonmutagen is a chemical or
substancethat doesnotdemonstrateevidence of one of the structural alerts associated with electrophilic potential
and that the chemical does not induce mutations in the Salmonella assay nor induce chromosomal effects when measured in vivo(eitherinduction ofchromosome aberra-tionsormicronuclei).Althoughother mechanisms of geno-toxicity or mutagenesis exist, for example, interference with chromosomal metabolism or the mechanics of chro-mosomesegregation,therehave beennoassaysidentified
yet that arecapableofresolvingthosespecific properties
operational definition that we use, therefore, may miss someproportion ofincipient orindirectmutagens, butit provides the highest degree of specificity for carcinogen
identification. We believe that this operational definition
definesthechemicalgroupswith the highest probability of direct interaction with and damage to DNA. Chemicals that lack these properties comprise a very structurally
Iable 1. Nonmutageniccarcinogens.a Chemical Aldrin Allyl isovalerate 11-Aminoundecanoic acid Benzaldehyde Benzene Benzofuran Benzyl acetate Butyl benzyl phthalate C.I. Vatyellow4
Chlordane(technical grade) Chlorendic acid
Chlorinatedparaffins:C12,60%chlorine Chlorinatedparaffins: C23,43%chlorine Chlorobenzilate Chlorothalonil Cinnamylanthranilate Decabromodiphenyl oxide Di(2-ethylhexyl) adipate Di(2-ethylhexyl) phthalate 1,4-Dichlorobenzene(p-dichlorobenzene) p,'-Dichlorodiphenyldichloroethylene Dicofol N,N'-Diethylthiourea 1,4-Dioxane Furfural Furosemide Heptachlor Hexachloroethane Hydroquinone Isophorone d-Limonene
Malonaldehyde, sodium salt Melamine
Nitrilotriacetic acid(NTA) N-Nitrosodiphenylamine Pentachloroethane Pentachlorophenol Phenylbutazone Piperonylsulfoxide
Polybrominated biphenyl mixture (Firemaster FF-1) Reserpine 2,3,7,8-Tetrachlorodibenzo-p-dioxin 1,1,1,2,-Tetrachloroethane 1,1,2,2-Tetrachloroethane Tetrachloroethylene 1,1,2-Trichloroethane
aAll chemicals listed are negative for structural alerts and were
Anotherline of evidence thatsupports a nonmutagenic originofsome cancers canbe derived fromthe occurrence
ofspontaneous tumors. Virtually all mammalian species
have demonstrated evidence of tumors when older
indi-vidualshave been examined. The best dataonthe incidence and patterns of spontaneous tumors are derived from studies with inbred mouse and rat strains. Among the
mostcomplete data availablearethose derivedfromthe 2 yearrodentbioassays conducted bythe National Toxicol-ogyProgram[NTP(10,11)]. Intheprotocolusedto assess
carcinogenicityin rodents, thereare concurrent controls of50miceorratsof each sexthatareheld fora104-week
exposure period and subsequently undergo complete
postmortem evaluation. The thousands of animals that
have been studied have demonstrated fairly consistent patterns of spontaneous tumor development, which has been maintained over manygenerations. Both mice (i.e., B6C3F1 hybrid) and rats (i.e., F344 strain) are housed
underhighly controlled conditions, andthedietsthey are
fed are well characterized and contain known, but extremelynegligible,amounts ofpotentially carcinogenic substances. The highly defined and controlled environ-mentprovides few,if any, sourcesofcarcinogens.Thus, the
constancyof thepatternoftumordevelopment within each
sexand species indicatesthat there areparticular genetic determinants that are responsible for theoccurrence of
The actual frequency of tumors developed at certain sitesdofluctuateand overtime;for example, the incidence
of mammary tumors or leukemia in rats has tended to increase. The increase in these latter tumors has been associated with improvementinmaintenance conditions,
could decrease the health of the animals and also attributed to improved dietary conditions that result in
relatively high weightgain (11). Thus, there are
environ-mental factors that can influence the incidence of spon-taneous tumors but do not significantly influence the pattern with which these tumors develop. The origin of spontaneoustumorsis unclear. Aspecific genetic influence hasidentified thehigh frequency of livertumorsoccurring
B6C3F1mice that isattributedto alocuscalledHcs(12). The genetic basis ofother types of spontaneous tumors
has not been well studied, but crosses between strains
showing high tissue-specific tumor incidence and other strains showing low tumor incidence at the same site, generally results in an intermediate level of tumor
expression in the F1 progeny,
suggestingthat in most casestheexpressionof spontaneoustumorsisdominantor
Among various ideas proposed to account for spon-taneoustumorigenesisisthe concept of DNAdamage of
endogenous origin. That is, mutations that could occur either as a consequence ofmistakes in DNArepair and
replicationmechanisms(14)orfromdamagethatoccurs as a consequence of normal metabolism
variousradicals ofoxygen,such assuperoxideorhydroxyl,
aregenerated (15-17). However, it is difficultto reconcile
these hypotheses with the spontaneous tumor patterns that develop inthe B6C3F1 mice and F344 rats because
neither can account for the tissue specificity of spon-taneous tumor incidence. If generalized DNA replication
orrepair errors oroxidative damagewereresponsible for the spontaneous tumors, onewould expect a more
gener-alized pattern of spontaneous tumor development that would be relatedtoeither'the tissues with the highest level ofendogenous cellular proliferationor tothose tissues that
have the highest levels of endogenous oxidative metabo-lism.Thepatternoftumorsobserved donotreflectthese patterns(10),andothermechanisms by which suchtumors
could arise should be considered.
It also has been proposed thattissues inwhich spon-taneous tumors arise undergo a higher level of spon-taneous initiation and that the action of nonmutagenic carcinogens may involve only the promotion or clonal expansion of suchspontaneously initiated cells. To evaluate this hypothesis, we have looked at the sites of
tumori-genesis that have been associated with exposure to 154
(9). As shown in Figure1, the carcinogens have induced
tumorsinabout30 tissuesites,butthe majority of chemi-calscanbe foundtobeactiveatapproximately10different
sites. These frequencies of induced tumorigenesis were thencomparedtothe sites of spontaneous tumorigenesis demonstrated in control animals. One example, male
Fischer rats (Fig. 2), shows a high level ofspontaneous
200 F 180 120 100 0
C,8 o 8 0 0 *V 60
tumorigenesisinthe hematopoieticsystem,whichis asite of relativelyhighcellproliferation. However,the hemato-poieticsystemiscomparabletothe adrenal glandinboth
spontaneousandinducedtumorincidence,but the level of endogenous cellularproliferation issignificantly lowerin
adrenal tissue. Another site of relatively high endogenous cell proliferationisthe skin, which showssignificantly less
spontaneous tumorigenesis andwas not asite oftumors
induced bynonmutagenic carcinogens.Sites ofthe highest
levels of inducedtumors were the liver and kidney, which
differconsiderablyinboth levels ofendogenous oxidative metabolism and spontaneous tumorincidence.
Theseresults,therefore,suggestthefollowing: a)notall sites inthebodyareequallyatriskfor chemical-induced tumorigenesis, b) the sites of induced tumors are not
directly related to the level ofendogenous cellular
pro-liferation,c)thespontaneous patternsof tumorigenesisdo not appear to dramatically influence the sites oftumor
induction by exogenous chemicals and, d) the genetic
influence on spontaneous tumors appears tobe the
pre-dominant factorintheirexpression. Ifindirectmutations
of oncogenes are involved in the development of spon-taneous tumors, the source ofthose spontaneous muta-tions is also unclear, and relatively little information is
available. The only extensive studies that have been
con-ducted involve the liverinthe B6C3F1 mouse.Spontaneous
mutations involving the 12th or13th codon of the v-H-ras genehave beenidentifiedinmany spontaneousandinduced
tumors (12), but mutated or translocated forms of other
_ Male p Rat _ Female Rat Male Mouse _Female Mouse
FIGURE 1.Occurrence oftumorsitesinducedby154carcinogenstestedby the NTPaccordingtosexandspecies. See Ashbyand Tennant(9)forkeyto
S LU CS MS IIC SP BD HG N PTG K L HS P C LS E PAG OST
ZG SB CG SK U TV 0 PG MG IS UB TG AG SV UT B OC MT
CIRCULATORY KIDNEY Spontaneous SI Nonmutagenic Carcinogens LIVER LUNG MAMMARY SKIN(S.C.) PANCREAS PREPUTIAL | THYROID PITUITARY ADRENAL
FIGLURE 2.Thefrequencies of inducedtumorigenesisbynonmutagenic carcinogenscomparedtothespontaneous ratesfor controlanimals.
The theories proposed to account forthe carcinogeni-city ofnonmutagenic chemicalscanbe combined intotwo
major groups: indirect mutagenesis and altered gene
expression. They arenotmutually exclusive mechanisms.
It isverypossible thatsomechemicals involvea
combina-tion of the two mechanisms, and there are even data to
suggest thatamong the mutagenic carcinogens, indirect
mutagenesis oralterationsintheexpression of important
target genes can be critical components in the
carcino-genic processes (19).
Currently,themostintensespeculationabout nonmuta-genic carcinogensconcerns therelationship between tox-icity, sustained tissue damage, and induced cellular proliferation ormitogenesis (20).Cellproliferation
gener-ally referstocompensatoryorreparative celldivisionthat is a consequence of toxicity, whereas mitogenesis
gener-ally refers to the capacity ofachemical or substance to
directly elicit cell division. Investigations by Totter (15), Ceruti(16),and Amesand Gold(17)haveproposedthat by-products of the normal oxidative metabolism of cells gives rise to relatively high levels of free radicals such as
superoxide orhydroxyl that have the capacityto damage
DNAand to induce mutations. Ames and Gold (17) have
focused on chemicals that induce toxicity and suggest that reparative processes associated with toxic injury such as the infiltration ofmacrophages canincrease the level of
oxyradicals. They propose that toxic injury sustains cell proliferation and can promotethe development of tumors
by providingfor the clonalexpansionof cells damaged by oxyradicals. Evidence in support of this mechanism has been offeredbytheidentificationof8-hydroxyguanosine.
This is an altered DNA base that occurs as a result of
oxyradical-induced DNA damage. However, there is no way to determine whether such altered bases occur in healthy cells that are dividing and have the capacity to repair suchdamageorwhethertheyoccurpredominantly
in cells that are irreversibly injured by toxicity and thus could not contribute to eithertheproliferativeprocessor to subsequent developmentoftumors.
Asecond mechanism by which endogenous sources of DNAdamage could arise has been summarized byLoeb (14). He has proposed that lesions induced in DNA by
mistakes inreplicationandrepairprocessesinduce spon-taneous mutations and that such mutational events could accountforasignificant proportionofendogenously initi-ated cells.
amplificationorclonal expansion ofendogenousDNAdamagedcellsormutated cells (the "mitogenesis and mutagenesis" theory) come from two lines of evidence. The first is derived from an extensive evaluation ofanumber of chemicals that induce
Toxicitythat issustained for much of the 104-week period of chemical exposure in these bioassays can result in
specifictissues.Such changescanoccurintheabsence of
neoplasia (19). The bioassays are carried out for 2 years,
whichrepresentsapproximately60%of the lifespanof the animal. The animalsundergo completepostmortem exam-ination at theend ofthebioassay period, andthere is little
basis for arguing that tumors might be detected ifthe injury was sustained longer or if the animals were
The second line of
evidenceisbased on the pattern of spontaneous tumorsthatoccursinthemiceandratsused
inthe bioassay. As discussed previously, the data do not support generalized induction ofspontaneously initiated cells either by errors in proliferation and repair or by oxidativedamage. Comparisons of the sites of spontane-ous tumorswiththesitesofhighest levelsofendogenous cellularproliferationorof oxidativemetabolism show very
little relationship to the pattern ofspontaneous tumors. Therefore, it appears unlikely that the spontaneously occurringtumorsinrodentscanbedirectly accounted for by indirect mutagenesis mechanisms.These datado sup-portthe conceptthat the spontaneous tumorsarise as a
result of specific genetic determinantsintheanimal. This does not exclude thepossibilitythat somechemicals may
have the capacitytoinduce indirect oxidative damageor to increasethe level of cellular and repairproliferation
mis-takes, andtosubsequently clonallyamplify these mistakes whentheyoccurinanoncogeneorsuppressor gene.Itis a plausible mechanism for some tumorigenesis, but it is
unlikely to account for the full range of carcinogenic effects observedamongthelarge numberofnonmutagenic carcinogens.
Ifthe clonal amplification ofspontaneous orindirectly initiated cells is not a common mechanism of
carcino-genesis, then howmay
tumors? Thereare different linesofevidence thatcanbe
.supportarole for alterationsinthe expres-sion of one or more of the critical proto-oncogenes or proto-suppressor genes, that is, theendogenous formsof
theoncogenes or suppressor genesplay critical rolesinthe complex regulatory pathways that control normal cell
functions.Ananalogycanbe drawn from theprocessesof differentiation wherein sequential changes in the
expression ofgenes andthe responses ofdifferentiating
cellstothe geneproductsresultinheritablealterationsin the patternofgeneexpression. Suchchangesareacquired by,and oftenmodifiedin,progenycells.When appropriate
stages of differentiation arereached,the patternofgene
expression canbecome fixed and
subsequentlyinherited by daughter cells arisinginthose tissues.Therefore,it is
possiblethat somechemicalsorenvironmentalagentscan actby
alteringtheexpression of thesecriticalregulatory
genesandgiverisetoprogenycells in which theheritable phenotypic change provides a growth advantage. The
development of tumors in response to so-called
"solid-state" carcinogens, such as plastic strips, films, or calculi
could involve suchamechanism. The cells adapttogrowth
inthepresenceof, or on the foreign objects. Progressively, moredysregulatedcells can emerge with a growth advan-tagethateventuallybecomesa tumorphenotype(21). The
question of whether mutations are induced in critical targetsin such cells has notbeen addressed.Therefore,it is inadequate to assume that mutations are required to
elicitneoplasticgrowth under these conditions.
Hormonal carcinogenesisprovidesasecond line of
evi-dence.The profound changes on normal regulatory pro-cessesinduced by hormones involve complex interactions with surfaceorintracellularreceptors and the transduc-tion of signals to the nucleus where changes in gene
expressionareaffected by theactionofvarious
transcrip-tion factors. Oncogenesthatplayroles in these processes have beenidentified byvirtueof the mutatedforms of the genesthat existinandweretransducedbyretroviruses. However,there arenormal cellularcounterpartsforsuch genes. For example, the erbA protein functions as an intracellular receptorforthyroid hormone (T3) and func-tions as a negative regulator of transcription (22). The
mutatedform of erbA has beendemonstratedtoplay a role in tumorigenic processes. However, constitutive expres-sionof the endogenousc-erbA proto-oncogenecan plausi-blyresultin similar events. Forexample, ifachemicalcan
functionasaligand for the thyroid hormonereceptorand significant levels of the chemical are present for
pro-tracted periods of time, it is possible that the normal regulatory functions ofthe receptor willbe subverted and
thatthe erbAgene product would be constitutively
pro-duced. Theconsequencecouldbeadysregulatedpatternof cellular proliferation because there is selection for more
rapidlyproliferating cell populations. Daughter cells also requiredtoexist inthepresenceofthechemical would also possess the altered phenotype. Such
pro-liferating cells would provide a fertile environment in whichsubsequent genetic changes couldoccurand leadto amalignantphenotype. Thus, the proliferationofthyroid cellsmaybefundamentally differentinthepresenceof a
nonmutagenicchemical that canalterthe process of gene
Pathways suchasthatproposed for the erbAoncogene
provideabasis for viewing the emergence of some cancers as an adaptive process. In this hypothesis it is not the direct actionofthechemical thatinduces
in cells, but rather that the chemical elicits adaptive
responses on the part of cells that lead to deregulated growthpatternsandtheemergenceof neoplasticvariants
(21).While this may seemtobeaminordistinction between
the actions of somenonmutagenic chemicals,ithas
impor-tantimplications. Forexample,the adaptiveprocess may
beintrinsicallymorereversiblethan the inductive process. In the absence of the chemical, reversion to a normal
phenotypemaybepossible. Numerousexamples of rever-sion orremodeling have beenseeninstudiesof hepatocar-cinogenesis andintheneoplastic transformationof cells in
are aproduct thatcanbe studiedandquantitated farmore readily than can changes in the patterns of gene tran-scription. It is necessary to explore further the complex
molecular interactionsoftranscription factors with DNA
binding sites and to determine ifspecific chemicals can
dysregulate the expression of critical control genes in ways that do notinvolve changesin DNA sequence (23).
Thus, I am proposing that at least some portion of
cancers are diseases of transcription that arise through mistakesinthe complexprocessof transcriptional
regula-tion and that some chemicals have the capacity to illicit such changes independent of their ability to stimulate
cellularproliferation. Induced cellular proliferation, there-fore,maybeanessentialcomponentallowingforthe clonal amplification of transcriptionally altered cells. However,
chemicals that can induce cellular proliferation directly throughamitogeniceffect,orindirectlythrough elicitinga compensatory response tothetoxiceffects of the chemical,
are not necessarily carcinogenic. Other properties ofthe
chemical, related to their ability to specifically interfere with the transcriptional process, may be the important property that distinguishes this class of nonmutagenic carcinogen from both other nonmutagens andmutagens that arecarcinogens.
1.IARC. Monographs on the Evaluation of Carcinogenic Risks to Humans. Overall Evaluations of Carcinogenicity, Supplement 7.
International Agency for ResearchonCancer, Lyon, 1987. 2.Cairns, J.The origin of humancancers.Nature 289:353-357(1981).
3.Henderson, B. E., Ross, R. K.,and Pike, M.C. Toward the primary preventionof cancer. Science254:1131-1137 (1991).
4.Boveri, T. TheOrigin of MalignantThmors.Williams and Wilkins, Baltimore,MD, 1929.
5.Solomon,E,Borrow, J., andGoddard, A.D.Chromosome aberrations andcancer.Science 254: 1153-1160 (1991).
6.Bishop, M. Molecular themes in oncogenesis. Cell64:235-248(1991).
7.Weinberg, R. A. Tumor suppressor genes. Science 254: 1138-1145 (1991). 8.Fearon, E. R., and Vogelstein, B. A genetic model for colorectal
tu-morigenesis.Cell 61: 759-767 (1990).
9.Ashby, J., and Tennant, R. W. Definitive relationships between chemi-cal structure, carcinogenicity and mutagenicity for 301 chemichemi-cals tested bythe US NTP. Mutat. Res. 257: 229-306 (1991).
10. Haseman, J. K.,Huff, J. E., Rao, G. N., Arnold, J. E., Boorman, G.A., andMcConnell, E. E. Neoplasms observedinuntreated andcornoil gavagecontrol groups of F344/Nratsand(C57B1/6N x C3H/Hen) Fl(B6C3F1) mice. J. Natl. Cancer Inst. 75: 975-984 (1985). 11. Haseman, J. K., and Rao, G. N. Effects of corn oil, time-related
changes, and inter-laboratory variability on tumor occurrence in control Fischer 344 (F344/N)rats.Toxicol.Pathol.20:52-60(1992). 12.Drinkwater, N. R.,and Ginsler, J. J. Genetic control of hepatocar-cinogenesis inC57B1/6J and C3H/Hev inbred mice. Carcinogenesis 7:1701-1707 (1986).
13. Eker, R.,and Mossige, J. A.Adominant gene for renal adenomasin
therat.Nature 189: 858-859 (1961).
14.Loeb, L. A. Endogenous carcinogenesis: molecularoncology into the twenty-first century-presendial address. Cancer Res.49:5489-5496
15.Totter, J.Spontaneous cancer and its possible relationshiptooxygen metabolism.Proc. Natl. Acad. Sci. USA 77: 1763-1767 (1980). 16.Cerutti, P., Larsson, R.,and Krupitza, G. Mechanismsof Oxidadant
Carcinogenesisin Genetic Mechanisms in Carcinogens and Tumor Progression.Wiley-Liss, Inc., 1990, pp.69-82.
17.Ames, B. N., and Gold, L. S. Too many rodent carcinogens: mitogenesis increasesmutagenesis.Science249: 970-971(1990). 18.Reynolds, S. H., Stowers, S. J., Patterson, R. M., Maronpot, R. M.,
Aaronson, S. A., Anderson, M. W. Activated oncogenes inB6C3F1 mouselivertumors:implications for risk assessment. Science 237: 1309-1316 (1987).
19.Tennant, R. W., Elwell, M. R.,Spalding, J.W,and Griesemer, R. A. Evidence that toxic injury is notalwaysassociated with induction of chemical carcinogenesis. Mol.Carcinog.4:420-440(1991). 20.Cohen, S. M., and Ellwein, L. B. Genetic errors,cell proliferation and
carcinogenesis. Cancer Res.51:6493-6505(1991).
21.Farber, E., and Rubin, H. Cellular adaptation in the origin and development ofcancer.Cancer Res.51:2751-2761 (1991).
22.Damm, K., Thompson, C.C., and Evans, R. M. Protein encoded by v-erb A functionsas athyroid-hormone receptor antagonist. Nature 339:593-597 (1989).
23.Issemann, I., and Green,S. Activation ofa memberof the steriod hormone receptor superfamilyby peroxisomeproliferators. Nature