Modulation of DNA adduct formation by successive exposures of DNA to small and bulky chemical carcinogens

The effect of a carcinogen—DNA adduct on the formationof a second adduct upon subsequent exposure to a second carcinogenwas studied using (i) a modified Maxam-Gilbert chemical sequencingreaction and (ii) a DNA synthesis termination analysis. A DNAfragment of known sequence was reacted with micromolar concentrationsof N-methyl-N-nitrosourea (MNU), (+)-r-7,t-8-dihydroxy-t-9,10-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene(BPDE), N-acetoxy-2-acetylaminofluorene (N-acetoxy-AAF), oraflatoxin B₁-8,9-epoxide (AFB₁ epoxide) singly or in successivedouble reactions. N-Acetoxy-AAF adducts were sensitive to theMaxam-Gilbert sequencing reaction for guanine; these adductssignificantly blocked the formation of BPDE-guanine adducts.Treatment of DNA with BPDE, however, did not inhibit subsequentformation of N-acetoxy-AAF adducts. DNA synthesis terminationanalysis suggested that methylation of guanine altered subsequentarylation of guanine by N-acetoxy-AAF and AFB₁ epoxide, andthat combined treatment inhibited replication, an effect notseen after MNU treatment alone. Possible mechanisms for themodulation of carcinogen binding are discussed. It is concludedthat not only is DNA damage by genotoxic carcinogens dependentupon both the chemical nature of the carcinogens and the nearestneighbors to target guanine bases, but that the effect of subsequentexposures to carcinogens is not always additive.     

Protective role of thiols in carcinogen-induced DNA damage in rat liver

Biological thiols are known to play an important role in thedetoxification of xenobiotics, including chemical carcinogens.To determine the influence of cellular thiols on carcinogeninduction of hepatic DNA damage in the rat, diethylmaleate (DEM)administration was used to deplete intracellular gluta-thione(GSH). The effects of administration of the synthetic thiols,N-acetylcysteine (NAC) and alpha-mercaptopropionyl-glycine (MPG),on the induction of DNA lesions were also examined. Pretreatmentwith DEM reduced liver GSH levels by >70%. As assessed bythe technique of alkaline elution, subsequent administrationof N-methyl-N'-nitro-N-nitroso-guanidine (MNNG) resulted inDNA damage 4 h post MNNG treatment which was 4- to 8-fold greaterthan that induced in the livers of rats treated with MNNG alone.However, DEM pretreatment had little effect on the extent ofDNA damage induced by methylnitrosourea (MNU). DEM alone didnot cause any measurable DNA damage. Pretreatment with MPG orNAC reduced MNNG-induced DNA damage by as much as 77%. In contrast,MNU-induced DNA damage was increased by MPG treatment whereasNAC treatment was without effect. These results indicated thatin the rat liver, the activity of some DNA alkylating agentsmay be modulated in varying degree by the concentration of intracellularthiols. These data support the notion that thiols play an importantrole in protection against carcinogen damage, and that syntheticthiols such as MPG and NAC may be useful as anti-carcinogenicagents against certain carcinogens.     

The O6-methylguanine-DNA activity of rat hepatoma cells is increased after a single exposure to alkylating agents

The O⁶-methylguanine-DNA-methyltransferase activity was measuredin rat hepatoma cells (H4 cells) at different times after N-methyl-N'-nitro-N-nitrosoguanidine(MNNG), methyl- methane sulfonate (MMS) or ethylinethane sulfonate(EMS) treatment. Incubation with MNNG (10µM) first depletesthe methyltransferase activity, then the number of methyltransferasemolecules per cell increases and reaches-3-fold the constitutivelevel after 48 h. Incubations with MMS (0.5 or 1 mM) or withEMS (5 or 10 mM) do not modify or partially decrease the constitutivemethyltransferase level. However, an enhancement of the activityis also observed after 48 h: the activity is 5-and 4-fold higherthan the control value in MMS-and EMS-treated cells, respectively.The methyltransferase increase is due to de novo protein synthesis.It is not observed in cells constitutively lacking this protein.The data suggest that the O⁶-methylguanine (O⁶-meGua) repaircapacity of H4 cells can be increased after a single treatmentwith alkylating agents, by a process different to the adaptiveresponse.     

Mechanism of N-acetylcysteine (NAC) and other thiols as both positive and negative modifiers of MNNG-induced mutagenicity in V79 Chinese hamster cells

Carcinogenic xenobiotics can be detoxified by nucleophilic thiols,which interact directly or through enzyme catalyzed reactionswith electrophilic metabolites/compounds or metabolically producedoxidants. Formation of such conjugates is assumed to be a dominatingcompetitive pathway reducing the mutagenic and carcinogeniceffects of several known carcinogens. In the case of the potentcarcinogen N-methyl-N'-nitro-N-nitrosoguanidine (MNNG) the situationis different since this carcinogen is transformed to reactiveintermediates by nucleophilic agents such as thiols. As a consequenceof this, the modulating effect of thiols has to be discussedin relation to the reaction kinetics of this nitroso compound.In this report we present data on the mutagenicity of MNNG whichwas enhanced by intracellular excess of glutathione, achievedby pre-treatment with N-acetylcysteine, cysteine and glutathione.Depletion of GSH by treatment with diethylmaleate or buthioninesulfoximineresulted in a decreased mutagenic effect of MNNG. A decreasedeffect of MNNG was also obtained by extracellular excess ofthiols. The modification of MNNG-mutagenicity was compared tothe effect of thiols on the mutagenicity of N-methyl-N'-nitrosourea(MNU) in V79 Chinese hamster cells. No effect of the thiolscould be detected on the mutagenicity of MNU, indicating thatthe intermediates of MNU and MNNG react, in agreement with thereaction kinetics, in favour of water and thiols become lessimportant. The results indicate that the activation of MNNGto mutagenic intermediates is occurring within the cells. Thisactivation is mediated by reaction of MJ with thiols (and amines)and subsequent release of short-lived alkylating intermediates.The mutagenic effect of MNNG can be reduced or enhanced respectivelyby decreas ing or increasing the intracellular thiol levels.As demonstra ted with MNU, intracellular thiol concentrations(in milli/molar range), which is high enough to bring aboutthe activation of MNNG, are not sufficiently high to protectDNA from damage by the alkylating intermediate. Extracellularlevels of thiols ‘protect’ the cells simply by increasingthe decomposition of MNNG in the treatment solution.     

Reactions of ultimate carcinogens with cell membranes: importance of carbamoylation of phosphatidylethanolamine by the carcinogens

The strong carcinogens, N-methyl-N-nitrosourea (MNU), Nmethyl-N'-nitro-A'-nitrosoguanidine(MNNG) react with pfaosphatidylethanolamine (PE) in chickenerythrocyte ghosts or rat kidney cells, whereas a weak carcinogen,methyl methanesulphonate (MMS), does not. The reactions of theseagents with commercial PE were also carried out and their reactivitieswere compared with their carcinogenic potencies. When comparedwith the amount of PE of the blank, 64%, 18% and 0% of PE werereacted with MNNG, MNU and MMS respectively. This indicatedthat MNNG is 4 times more reactive than MNU, while MMS doesnot react with PE, correlating with their tumorigenic potencieson mouse skin. A positive correlation was also observed betweenthe reactivity of a series of N-methyl-N-aryl-N-nitrosoureas(I-X)to PE and their tumorigenicity. In the case of the nitrosoureas,the reactions proceeded through carbamoylation of the aminogroup in PE by the isocyanates generated from the agents. Furthermore,many isocyanates also reacted with PE just as the strong carcinogensdid. Our present results suggest that both alkylation of DNAand reaction with cell membranes would be required for the formationof transformed benign cells. For progression of the benign tomalignant cells, further alkylation of DNA would be required,and the carcinogenic potency of the agents may result from acombination of both reactions.     

Increased susceptibility to in vitro transformation of cells carrying the Eker tumor susceptibility mutation

Rats carrying the Eker tumor susceptibility mutation are geneticallypredisposed to renal cell carcinoma. Rats heterozygous for theEker mutation (Eker carriers) develop multiple bilateral renalcell carcinomas by the age of 1 year. Using an in vitro ratkidney epithelial (RKE) transformation assay developed in ourlaboratory, proximal tubule cells derived from known Eker ratcarriers (+/ek) and non-carriers (+/+) were exposed to the carcinogenN-methy-N '-nitro-N-nitrosoguanidine (MNNG), to determine ifcells derived from Eker carriers were more susceptible to invitro transformation than cells derived from noncarrier animals.The percent transformation frequency following MNNG treatmentwas 7.5-fold higher in cells derived from carrier animals whencompared to cells from non-carrier animals. This increased susceptibilityto transformation due to inheritance of the Eker mutation isconsistent with a predisposition resulting from inactivationof a tumor suppressor gene. The increased susceptibility ofkidney epithelial cells carrying the Eker mutation may proveuseful in the further development of the RKE transformationassay as a sensitive tool to identify potential renal carcinogens.In addition, because transformation frequency in the RKE assaymeasures a very early step in multistage transformation, theseresults also suggest that alterations (by Loss of Heterozygosityor mutation) at the Eker tumor susceptibility locus are an earlyevent in the development of renal tumors in the rat.     

Mutagenesis and morphological transformation by N-methyl-N''-nitro-N-nitrosoguanidine in the BALB/3T3 clone A31-1-1 cell line

Concomitant induction of ouabain resistant mutations and morphologicaltransformation in the BALB/3T3 ClA31-1-1-c mouse embryo cellline were obtained after a 30 min treatment with the directalkylating agent N-methyl-N'-nitro-N-nitrosoguanidine (MNNG).Parameters affecting mutation frequencies were determined. Theoptimal expression time (48 h) for ouabain resistance was independentfrom the dose of the carcinogen. A linear dose-response relationshipfor mutation induction was found after treatment with increasingdoses of MNNG. The ratio of malignant transformation to mutationfrequencies induced by the short treatment with MNNG was foundto be within the same order of magnitude over a four-fold doserange. The development of a mutational assay for ouabain resistancein the BALB/3T3 ClA31-1-1-c cell line makes quantitative comparisonspossible between mutation and neoplastic transformation frequenciesinduced by chemical carcinogens in this single cellular system.     

DNA mutilation of the pepsin 1 gene during rat glandular stomach carcinogensis induced by N-methyl-N'-nitro-N-nitrosoguanidine or catechol

The methylation patterns of the rat pepsinogen 1 (Pg1) genein preneoplastic and neoplastic stomach lesions induced by genotoxicN-methyl-N'-nitro-N-nitrosoguanidine (MNNG) or the non-genotoxiccarcinogen catechol were investigated. Male WKY/Ncrj rats weregiven MNNG in their drinking water (50 mg/l) for 30 weeks or0.8% catechol throughout the experiment (60 weeks). MNNG inducedPg1 altered pyloric glands (PAPG), adenomatous hyperplasiasand well-differentiated adenocarcinomas. Catechol also inducedPAPG and adenomatous hyperplasias although cancers did not develop.Adenomatous hyperplasias and adenocarcinomas all consideredof gastric type cells resembling surface mucous cells or pyloricgland cells with little or no Pg1 expression. In MNNG-inducedstomach cancers generally lacking Pg1, altered Pg1 gene methylationwas observed with both CCGG and GCGC sites being methylatedmore than normal pyloric mucosa. MNNG or catechol-induced adenomatoushyperplasias also demonstrated essentially the same methylationchanges in the CCGG, but not in the GCGC sites. In the mucosacontaining PAPG in groups treated with MNNG or catechol themethylation patterns of the Pg1 gene were quite similar to thoseof normal pyloric mucosa, although the CCGG sites tended todemonstrate slightly increased methylation. The results suggestthat the altered methylation of the Pg1 gene observed in stomachcancers is acquired early in the carcinogenic process and progressivemethylation changes occur with tumor development.     

Fluorescence and mass spectral evidence for the formation of benzo[a]pyrene anti-diol-epoxide--DNA and --hemoglobin adducts in humans

Highly specific methods are required to detect and quantitatecarcinogen—macromolecular adducts in humans who are exposedto complex mixtures of chemical carcinogens. High performanceliquid chromatography and fluorescence spectroscopy have beenused successfully to detect and identify residues of benzo[a]pyrene-7,10/8, 9-tetrahydrotetrol (BP-7, 10/8, 9-tetrol) that were releasedupon mild acid hydrolysis of human DNA or hemoglobin. Synchronousfluorescence spectroscopy data indicate that levels of benzo[a]pyrene-diol-epoxide—DNA(BPDE—DNA) adducts as high as 1.54 fmol BPDE/µgDNA are formed (1 adduct in 5 million nucleotides) in peripheralblood lymphocytes of coke-oven workers; these data were subsequentlycorroborated by gas chromatography/mass spectroscopy singleion monitoring analysis (m/z 404+). Additionally, among lungcancer patients, 5 samples of tumor DNA were found to be negativeand 1 of 4 samples of corresponding lung tissue was found tobe positive. Extraction and purification of BP-7, 10/8, 9-tetrolfrom the hemoglobin of smokers suggested levels of bound carcinogenin excess of 1 ng BPDE/gm of hemoglobin. High performance liquidchromatography combined with synchronous fluorescence spectroscopyprovides a highly specific method for the detection of covalentlybound BP residues in both human hemoglobin and DNA.     

Chromosomal and c-K-ras oncogene alterations induced by a chemical carcinogen and phorbol ester in skin fibroblasts of individuals with familial polyposis coli

Skin fibroblasts derived from three patients with familial polyposiscoli (FPC) were treated in vitro with N-methyl-N’-nitro-N-nitrosoguanidine(MNNG) alone or in combination with the tumor promoter, 12-O-tetradecanoylphorbol-13-ace-tate(TPA). None of the cultures treated four times with MNNG alone(1µg/ml each) or in combination with TPA (eight applications,0.1 µg/ml each) showed either morphological transformationor anchorage-independent growth for 18 months after treatments.FPC cells treated with MNNG alone showed cell growth inhibition,breakage and loss of chromosomes, as well as the deletion of5.7 kilobase (kb) EcoRl fragment in the c-K-ras locus as detectedby Southern blot analysis with v-K-ras specific probe (cloneKBE-2). The control cells contained two EcoRl fragments of 5.7and 4.2 kb. On the other hand, cells treated with MNNG firstand then followed by treatment with TPA not only showed an increasein the rate of cell growth but also exhibited two novel EcoRlfragments of 9.4 and 12 kb. Treatment with TPA alone appearedto stimulate cell division long after application and to inducechromosomal aberrations but had no effect on c-K-ras sequences.The most likely explanation for the appearance of chromosomepulverization and hyperploidy in FPC cells treated with TPAis the induction of premature chromosome condensation of theS phase cells due to fusion with the mitotic chromosomes.     

Carcinogen-induced unscheduled DNA synthesis in xenotransplanted human tracheobronchial epithelium: comparison with rat tracheal epithelium

Extrapolation from rodent genotoxicity data to humans is complicatedby variables such as interspecies differences in carcinogenmetabolism and DNA repair. A xenograft system containing humanbronchial epithelial cells was used to assess the inductionof unscheduled DNA synthesis (UDS) by carcinogens and to comparethe response with that of rat tracheal epithelium. Cells fromhuman bronchus were grown in explant culture, inoculated intode-epithelialized rat tracheas and implanted subcutaneouslyinto nude mice. Within six weeks, a differentiated mucociliaryepithelium lined the xenografted tracheas. Fresh rat tracheasand human xenografts were cut into rings and incubated in mediacontaining [³H]thymidine and either the direct-acting car cinogen,N-methyl-N'-nitro-N-nitrosoguanidine (MNNG, 3–1000 µM),or a carcinogen requiring metabolic activation, 4-nitroquinoline-1-oxide(4-NQO, 3–100 pM). Tissues were then fixed, sectioned,processed for autoradiography and the number of nuclear grains(NG) determined for 100 epithelial cells lining the tracheain each section. A time- and concentration-dependent increasein NG was observed in both human xenografts and rat tracheasafter treatment with MNNG or 4-NQO, indicating induction ofUDS by these agents. The UDS response to MNNG in the human xenograftswas similar to that observed in the rat tracheas, whereas theresponse to 4-NQO was greater in rat tracheas. These studiesindicate that the human xenograft system should have applicationsfor the study of carcinogen-induced damage in normal human bronchialepithelial cells.     

Identification in rat stomach mucosae of a cell population characterized by a deficiency for the repair of O6-methyldeoxyguanosine from DNA

Immunohistochemical studies have been used to show the timecourse for the cell-specific methylation of DNA in the uppergastrointestinal tract of rats treated with N-methyl-N'-nitro-N-nitrosoguanidine(MNNG). The doses used were 1, 5, 25 and 50 mg MNNG/kg (i.g.)and tissue samples were analysed at intervals of from 1 to 192h after treatment. Relatively little reaction with nuclear DNAwas observed in the forestomach and still less in the oesophagus.Reaction with DNA was most extensive in the corpus, pylorusand the duodenum, reaching a peak of staining intensity between2 and 4 h and then declining progressively thereafter. Stainingfor the presence of O⁶-methyldeoxyguanosine (O⁶-MedG) in DNAwas highly selective and tended to occur in the nuclei of thebasal cells of the oesophagus and forestomach and in the cellsof the lumenal border of the glands and villi of the corpus,pylorus and duodenum. There were also areas, 5–15 glandsapart where staining for O⁶-MedG in the corpus and pylorus extendedas far down as the basal mucosa. From 12 h after MNNG treatment,in the corpus and pylorus, a band of strongly methylated cellsbecame apparent about 3–6 cells deep from the lumen andremained identifiable up to 168 h after treatment with the higherdoses. These cells, which apparently have a very low O⁶-MedGrepair capacity, are stationary (i.e. not part of the escalator)and are located in the mesenchymal tissue elements as demonstratedby staining of serial sections with cytokeratin or vementin.The significance of this population of cells is unknown.     

Inhibition of DNA synthesis, independent of DNA adduct formation, by benzo[a]pyrene diol epoxide in mammalian cells

Treatment of SV40-infected CV-1 cells with the ultimate carcinogenanti-benzo[a]pyrene diol epoxide (BPDE) at 1 x 10 ^–4 mg/mlor higher reduced the rate of viral DNA synthesis to an extentdependent on the BPDE concentration; similar reductions in cellularDNA synthesis were produced in infected and uninfected cells.Treatment of cells with BPDE, followed by removal of BPDE, atvarious times before infection with SV40 gave the same results.Recovery or partial recovery of DNA synthesis occurred whenthe BPDE concentration was below 6 x 10 ^–4 mg/ml; at higherconcentrations the cells were killed. Simultaneously replicatingviral DNA's from viruses infecting the same cells before andafter BPDE treatment of the cells exhibited the same reducedrate of synthesis. The evidence indicates that covalent adductsin viral DNA are not responsible for its reduced replicationrate; moreover, it is probable that an insignificant numberof adducts is produced in intracellular viral DNA at BPDE concentrationsthat do not kill CV-1 cells. Rather, it appears likely thatBPDE inhibits viral DNA synthesis by attacking cellular DNAor non-DNA targets. Caution is therefore required in relatingthe effects of BPDE and other carcinogens to DNA adduct formation.     

Para-methoxyphenol strongly stimulates cell proliferation in the rat forestomach but is not a promoter of rat forestomach carcinogenesis

The modifying effects of para-methoxyphenol (PMP) second stagetreatment on N-methyl-N'-nitro-N-nitrosoguanidine (MNNG)-initiatedrat forestomach carcinogenesis were investigated. Groups of15 6 week old male F344 rats were given a single intragastricadministration of 150 mg/kg body wt MNNG and starting 1 weeklater were administered powdered diet containing 2.0, 1.0, 0.5,0.25 or 0% PMP until they were killed at week 52. PMP causedepithelial damage and hyperplasia in a dose-dependent mannerin the fore-stomach epithelium, but nevertheless was not associatedwith any increase in the incidences of either papillomas orsquamous cell carcinomas. The results thus clearly indicatedthat stimulation of cell proliferation does not necessarilycorrelate with promotion in the second stage of two-stage forestomachcarcinogenesis.     

MNNG-induced partial phenotypic reversion of Mer- cells

The effect of pretreatment with N-methyl-N'-nitro-N-nitrosoguanidine(MNNG) on MINNG sensitivity of the surviving population wascompared in two HeLa lines, one of the Mer⁺ phenotype (HeLaS3) and one of the Mer^– phenotype (HeLa MR). Whereas MNNGpretreatment of HeLa Mer⁺ cells had no effect on the MNNG sensitivityof surviving cells, Mer^– cells surviving a first exposureto MNNG became much more resistant to MNNG Comparison of thesensitivity of individual HeLa MR clones with their MNNG-pretreatedpopulation and analysis of the composition of the pretreatedpopulation showed that the majority of cells surviving the MNNG-pretreatmentnow displayed the Mer⁺ phenotype in respect to sensitivity toMINNG. One MNNG-resistant clone derived from a pretreated HeLaMR population (Cl 4) was characterized further. It had a similarsensitivity to the Mer⁺ line to all monofunctional alkylatingagents, but was as sensitive as the Mer^– line to the crosslinkingagent chloroethylntrosourea. Cl 4 cells, like the Mer^–cells, did not repair O⁶-methylguanine (O6MeG). The resultssuggest that the two characteristics which are usually coupledwith the Mer^– phenotype - lack of O⁶MeG repair and hypersensitivityto MNNG - can be separated.     

Dosimetry of PAH skin carcinogenesis: covalent binding of benzo[a]pyrene to mouse epidermal DNA

Benzo[a]pyrene (BaP) is metabolized to the chemically reactiveanti and syn isomers of the 7,8-diol-9,10-epoxides of BaP (BPDE)which bind covalently to DNA to form DNA/BPDE complexes. Tetrolsliberated from the DNA/BPDE complex by acid hydrolysis are easilyquantified by h.p.l.c. using fluorescence detection. This approachallows femtomole amounts of BPDE associated with the DNA isolatedfrom a single mouse to be detected using conventional instrumentation.The usefulness of this technique to estimate the interactionof BaP with DNA of mouse skin, both in the intact animal andin organ culture, was investigated. With mouse skin in organculture it could be demonstrated that: (i) upon a single topicalapplication of 5 µg of BaP, binding to DNA occurred viaBPDE at a linear rate for up to 65 h, (ii) the amount of bindingwas dose dependent at concentrations of BaP <10 µg.     

Evidence for cytosolic glutathione transferase-mediated denitrosation of nitrosocimetidine and 1-methyl-2-nitro-1-nitrosoguanidine

Nitrosocimetidine (NC) and 1-methyl-2-nitro-1-nitrosoguan-idine(MNNG) are closely related N-nitrosamidines. NC is the nitrosatedderivative of cimetidine (Tagamet), an orally administered compoundused extensively in the treatment of gastric ulcers. MNNG isa potent carcinogen capable of initiating tumors close to thesite of administration and used experimentally to produce stomachcancer. It has become evident that the primary metabolic fateof both of these agents is denitrosation. We have discoveredan activity in the cytosol fraction of hamster liver which iscapable of denitrosating these nitrosamidines with an efficiencyapproaching 100%. The activity is heat sensitive and requiresreduced glutathione as a cofactor. Inhibition of the denitrosatingactivity with compounds which inhibit in parallel the conjugationof glutathione with 1-chloro-2,4-dinitrobenzene (CDNB) providesevidence that the activity is glutathione transferase. One moleculeof reduced glutathione is consumed in each denitrosation event.Nitrite is formed as denitrosation proceeds with a yield equivalentto 25–50% of the denitrosated product produced. Glutathionedisulfide is a minor reaction product, representing 3% of thedenitrosation product yield in the MNNG case and 12% in theNC case. Thus far in our survey of N-nitrosamines, N-nifrosamidesand N-nitrosamidines, only the nitrosamidines appear to be vulnerableto the cytosolic denitrosating activity. In an attempt to evaluatethe importance of the glutathione-dependent reaction in theintact hamster, we have depleted glutathione by pretreatmentwith the commonly used agents diethyl maleate (DEM) and L-buthionine-S,R-sulfoximine(L-BSO). Nitroso compound was administered i.v. and the circulatingblood levels of intact and denitrosated compound 5 min afterdosing quantified. NC-and MNNG-derived methylation of organDNA was also monitored. Pretreatment had no effect on the cytosolicdenitrosating or CDNB-conjugating activities. L-BSO pretreatmenthad no apparent effect on the denitrosative metabolism of NCor MNNG. With DEM pretreatment we obtained clear indicationsof a decreased rate of denitrosation and observed a 10-foldincrease in MNNG-derived liver DNA methylation. The differentialeffects of these pretreatments are taken as an indication thatDEM-sensitive proce other than those requiring glutathione dominateN-nitrosamidine denitrosation in the hamster.     

Arteriosclerotic plaque development is 'promoted' by polynuclear aromatic hydrocarbons

In previous work we found that weekly injections of the polynucleararomatic hydrocarbon (PAH) carcinogen 7,12-dimethylbenz[a]anthracene(DMBA) induced spontaneous aortic plaques in cockerels to growto a larger size and at a faster rate than plaques in controlanimals. To determine whether plaque-stimulating ability isrelated to carcinogenic potency or mutagenicity we have nowtested a variety of agents, including PAH carcinogens, non-PAHcarcinogens and weakly carcinogenic PAHs. Cockerels were injectedweekly (from 4–20 weeks of age) with one of the followingcompounds: benzo[a]pyrene (B[a]P), benzo[e]pyrene (B[e]P), dibenz[a,h]anthracene (AH), dibenz [a,c]anthracene (AC), 3-methylcholanthrene(MCA), acetylaminofluorene (AAF), N-methyl-N,N'-nitro-nitrosoguanidine(MNNG) or anthracene (ANT). Plaques were present in the abdominalaortas of all animals. Plaque volumes were 8–14 timesgreater in AC-, B[a]P-, B[e]P-, MCA- and AH-treated cockerelsthan in controls. Plaques were slightly larger in the AAF-treatedgroup than in control animals, and in the ANT- and MNNG treatedgroups were indistinguishable in size from plaques in controlanimals. The largest plaque volumes were in AH treated cockerelsand were comparable in size to those elicited by DMBA treatment.The accelerated development of plaques is consistent with a‘promotional’role for these agents. There was a poor correlation betweenmutagenicity or carcinogenicity and plaque ‘promotion’,which may reflect a role for different metabolites in theseprocesses.