Detection of activated proto-oncogenes in N-nitrosodiethylamine-induced liver tumors: a comparison between B6C3F1 mice and Fischer 344 rats

DNA from B6C3F₁ mouse and Fischer 344 rat liver tumors inducedby N-nitrosodiethylamine (DEN) were examined for the abilityto induce morphological transformation of NIH3T3 cells. DNAsfrom 14 of 33 of the mouse liver tumors induced by a singleinjection of DEN at 12 or 15 days of age were positive in thisassay while DNA from only one of 28 DEN-induced rat liver tumorswas active. Southern blot analysis of the NIH3T3 transformantsderived from the mouse liver tumors revealed amplified and/orrearranged restriction fragments homologous to the H-ras proto-oncogene.DNA from two independent foci induced by the rat tumor DNA didnot hybridize to probes for members of the ras gene family orc-raf. Activating mutations in the H-ras genes from the DEN-inducedmouse liver tumors were characterized by selective oligonucleotidehybridization and the detection of a new XbaI restriction siteby Southern blot analysis. In activated H-ras genes from theDEN-induced mouse liver tumor DNA, seven of 14 had a CGAT transversionat the first base of the 61st codon, three of 14 had an ATGCtransition and four of 14 had the ATTA transversion at the secondbase of codon 61. This spectrum of mutations is very similarto that recently observed in activated H-ras genes found inspontaneously occurring B6C3F₁ mouse liver tumors. Taken together,the data suggest that the DEN-induced rat and mouse liver carcinogenesismay involve genetic targets other than or in addition to theH-ras gene.     

Trans-1,2-dihydro-1,2-dihydroxy-6-aminochrysene is metabolized to form a major adduct with deoxyguanosine and produces mutations in the hprt gene of Chinese hamster ovary cells at G:C basepairs

6-Nitrochrysene can be activated to genotoxic derivatives bytwo major metabolic pathways: nitroreduction to N-hydroxy-6-aminochrysene,and a combination of ring-oxidation and nitroreduction thatinvolves the intermediate formation of trans-1,2-dihydro-1,2-dihydroxy-6-aminochrysene(6-AC-1,2-dihydrodiol). The DNA adduct formed from this latterpathway was evaluated by reacting individual deoxynucleoside5'-monophosphates with 6-AC-1,2-dihydrodiol in the presenceof liver microsomal enzymes from 3-methylcholanthrene-pretreatedrats. Binding was greatest to deoxyguanosine monophosphate andthe major deoxyguanosine (dG) adduct co-chromatographed withthe single major adduct formed from the microsome-catalyzedreaction of 6-AC-1,2-dihydrodiol with DNA. In order to characterizethe mutational changes associated with the 6-AC-1,2-dihydrodiolpathway, we analyzed the mutational spectrum produced by 6-AC-1,2-dihydrodiolin the hypoxanthine-guanine phosphoribosyl-transferase (hprt)gene of CHO-K1 cells. cDNA was synthesized from the RNA of 286-thioguanine-resistant mutants, the hprt coding region amplifiedby the polymerase chain reaction, and the DNA products directlysequenced. Twenty independent primary mutations were found:12 G:C T:A transversions, three G:C C:G transversions, oneG:C A:T transition, one A:T T:A transversion, two –1frameshift mutations in sequences containing consecutive guanines,and one 11 bp deletion. All G:C basepair substitutions had themutated dG on the non-transcribed strand and 86% of the G:Cbasepair substitutions had one purine 3' to the mutated dG.The pattern of 6-AC-1,2-dihydrodiol-induced basepair substitutionswas distinct from the pattern observed in solvent control mutants.These results are consistent with the formation of a promutagenicdG adduct from a metabolite of 6-AC-1,2-dihydrodiol.     

Utilization of 1, N6-Etheno-2'-deoxyadenosine 5'-triphosphate during DNA synthesis on natural templates, catalyzed by DNA polymerase I of Escherichia coli

To test whether vinyl chloride-induced mutagenesis might involveambiguous base pairing of 1, N⁶-etheno-adenine (A) during DNAsynthesis, we examined the base pairing potential of dATP duringDNA synthesis catalyzed by Escherichia coli DNA polymerase I(Klenow fragment). An electrophoretic assay of chain elongationwas used to assess the degree to which dATP could substitutefor each of the normal dNTPs during elongation of a primer annealedto a bacteriophage template. Despite the fact that the ethenobridge completely blocks normal Watson-Crick pairing of A withT, we observed that dATP could substitute for dATP during primerelongation (although inefficiently). In addition, detectablesubstitution of dATP for dGTP and dCTP occurred, indicatingthat A exhibits ambiguous base pairing properties. The relativeease of dAMP incorporation (opposite template T, C and G) appearedto vary considerably at different positions along the template.The major, form of eA incorporation (replacement of A) was confirmedby measurements of dATP-dAMP turnover (a commonly used methodfor detecting misincorporation), and also by the demonstrationthat A was present in enzymatic hydrolysates prepared from DNAthat was synthesized with dATP replacing dATP. A model for ambiguousbase pairing of dATP is proposed, in which incorporation occursvia the protonated, syn form of dATP.     

Correlation of (6-4)photoproduct formation with transforming mutations in UV-irradiated Ha-ras

Sites and types of mutations in relation to the amount of cyclobutanepyrimidine dimers (CPDs) and (6–4)-photoproducts in UV-irradiatednormal Ha-ras sequences were investigated. Mouse BALB/C 3T3cells were transfected with UV-irradiated pYN-mHras plasmidscontaining mouse normal Ha-ras sequences, and transformed focideveloped. Direct DNA sequencing of the Ha-ras retrieved fromthe foci revealed that most mutations (23/24, 96%) took placeat dipyrimidine sequences, and the CT transition at the 3'-cytosinein 5'-TC or 5'-CC sequences was predominant (17/24, 71%) incodons 12, 13 and 60. In codon 61, where 5'-TC or 5'-CC is absent,two mutations were found at the 5'-TT sequence. More (6–4)photoproductswere produced than CPDs in codons 12, 13 and 60, and more CPDswere produced than (6–4)photoproducts in codon 61. Theseresults suggest that (6–4)photoproducts are the majorlesion leading to the mutations in the mouse Ha-ras sequenceand subsequent transformation of BALB/C 3T3 cells.     

Completion of excision repair patches in human cell preparations: identification of a probable mode of excision and resynthesis

Excision repair of u.v. damage in human fibroblasts is moresensitive to inhibitors of DNA polymerase (cytosine arabinoside,aphidicolin) than to an inhibitor of polymerase ß(dideoxythymidine), which indicates a greater role in repairfor polymerase than for polymerase ß. These inhibitorsall generate shortened patches with free 3' termini; the detailedstructure of these patches was investigated in permeable cellsor isolated nuclei by degradation of DNA with exonuclease IIIand by resynthesis with DNA polymerase I (Klenow fragment) andT4 DNA ligase. The structure of the shortened patches appearsto be a short stretch of DNA synthesized in the 5'3' directionwithin a longer single-strand gap. The single-strand gap aheadof the 3' terminus can be bridged only by the combined actionof polymerase and ligase. This structure implies that excisionmust involve removal of an oligonucleotide or widening of agap by 5'3' exonuclease action to produce a single-strand regionwide enough to be a substrate for polymerase . There is no evidencefor structures generated by nick translation or strand displacement.     

Supplemental administration of 1{alpha}-hydroxyvitamin D3 inhibits promotion by intrarectal instillation of lithocholic acid in N-methyl-N-nitrosourea-induced colonic tumorigenesis in rats

The effect of 1-hydroxyvitamin D₃ [1(OH)D₃) on promotion byintrarectal instillation of lithocholic acid (LC) in N-methyl-N-nitrosourea(MNU)-induced colonic tumorigenesis was studied in a rodentmodel. Ninety-two female F344 rats received intrarectal injectionof 2.5 mg of MNU twice in one week followed by 1 mg of LC orits vehicle alone three times weekly for 48 weeks. Those whichreceived LC were given a concomitant intragastric administrationof 0.04 µg of 1(OH)D₃ or its vehicle alone three timesweekly. In the group receiving MNU alone (n=30) five rats borecolomc tumors; in the MNU + LC group (n=32) 15 and in the MNU+ LC + 1(OH)D₃ group (n=30) six rats bore colonic tumors (MNU+ LC versus MNU + LC + 1(OH)D₃ group, P<0.05). These resultsindicated that promotion of MNU-induced colonic tumorigenesisby LC was suppressed by supplemental administration of 1(OH)D₃.     

Development of monoclonal antibodies specific for 1,N2-ethenodeoxyguanosine and N2,3-ethenodeoxyguanosine and their use for quantitation of adducts in G12 cells exposed to chloroacetaldehyde

Monoclonal antibodies specific for N²3-ethenodeoxyguanosine(N²,3-dGuo) and 1,N²-ethenodeoxyguanosine (1,N²-dGuo) were developed.In a competitive ELISA, 50% inhibition of binding of the N²,3-dGuospecific antibody (ETH1) was achieved with 18 fmol of N²,3-dGuo.Fifty per cent inhibition of the 1,N²-dGuo-specific antibody(ETH2) required 11 pmol 1,N²-dGuo. Immunoassays for N²,3-dGuoand 1,N²-dGuo in single-stranded DNA were developed using theseantibodies. The immunoassays could detect as little as 48 fmolof N²,3-dGuo or 340 fmol 1,N²-dGuo in 25 µg of singlestranded DNA. These assays and previously developed immunoassaysfor 1,N⁶-ethenodeoxyadenosine (1,N⁶-dAdo) and 3,N⁴-ethenodeoxycytidine(3,N⁴-dCyd) were used to measure etheno adduct levels in DNAof cells exposed to chloroacetaldehyde. The cells used wereV79 cells with an inactivated hprt gene and a single copy ofthe bacterial gpt gene (G12 cells). The most abundant ethenoadduct was 1,N⁶-dAdo, followed by 3,N⁴-dCyd and N²,3-dGuo. 1,N²-dGuowas not detected in chloroacetaldehyde-treated G12 cells. Chloroacetaldehydewas also shown to be mutagenic in these same cells.     

1,N6-etheno-2' -deoxyadenosine and 3,N4-etheno-2'-deoxycytidine detected by monoclonal antibodies in lung and liver DNA of rats exposed to vinyl chloride

1,N⁶-Etheno-2'-deoxyadenosine (dAdo) and 3,N⁴-Etheno-2'-deoxycytidine(dCyd) are formed in vitro by reaction of DNA with the electrophilicmetabolites of vinyl chloride (VC), chloroethylene oxide andchloroacetaldehyde. To detect and quantitate these DNA adductsin vivo, we have raised a series of specific monoclonal antibodies(Mab). Among those, Mab EM-A-1 and Mab EM-C-1, respectively,were used for detection of dAdo and dCyd by competitive radioimmunoassay(RIA), following pre-separation of the etheno adducts from DNAhydrolysates by high perfonnance liquid chromatography. At 50%inhibition of tracer-antibody binding, both Mab had a detectionlimit of 187 fmol and antibody affinity constants (K) of 2 x10⁹ l/mol. The levels of dAdo and dyd were quantitated in theDNA of lung and liver tissue of young Sprague-Dawley rats exposedto 2000 p.p.m. of VC for 10 days. The dAdo/2'-deoxyadenosineand dCyd/2'-deoxycytidine molar ratios were 1.3 x 10^–7and 3.3 x 10^–7 respectively, in lung DNA, and 5.0 x 10^–8and 1.6 x 10^–7 in liver DNA. When hydrolysates of 3 mgof DNA were analyzed by RIA at 25% inhibition of tracer-antibodybinding, dAdo and dCyd were not detected in liver DNA from untreatedrats above the limiting dAdo/2'-deoxyadenosine and dCyd/2'-deoxycytidinemolar ratios of 2.2 x 10^–8 and 3.1 x 10^–8, respectively.     

Typical signature of DNA damage in white blood cells: a pilot study on etheno adducts in Danish mother-newborn child pairs

The impact of DNA damage commonly thought to be involved inchronic degenerative disease causation is particularly detrimentalduring fetal development. Within a multicenter study, we analyzed77 white blood cell (WBC) samples from mother–newbornchild pairs to see if imprinting of DNA damage in mother andnewborn shows a similar pattern. Two adducts 1,N⁶-ethenodeoxyadenosine(dA) and 3,N⁴-ethenodeoxycytidine (dC) were measured by ourultrasensitive immunoaffinity ³²P-post-labeling method. Thesemiscoding etheno-DNA adducts are generated by the reaction oflipid peroxidation (LPO) end products such as 4-hydroxy-2-nonenalwith DNA bases. Mean dA and dC levels when expressed per 10⁹parent nucleotides in WBC-DNA from cord blood were 138 and 354,respectively; in maternal WBC-DNA, the respective values were317 and 916. Thus, the DNA-etheno adduct levels were reliablydetectable and about two times lower in child cord blood, thedifference being significant at P < 0.0004. Analysis of dAand dC levels in cord versus maternal blood WBC showed strongpositive correlations (R² 0.9, P < 0.00001). In conclusion,LPO-induced DNA damage arising from endogenous reactive aldehydesin WBC of both mother and newborn can be reliably assessed bydA and dC as biomarkers. The high correlation of etheno adductlevels in mother and child WBC suggests that a typical signatureof DNA damage is induced similarly in fetus and mother. Prospectivecohort studies have to reveal whether these two WBC-DNA adductscould serve as risk indicator for developing hematopoietic cancersand other disorders later in life. Abbreviations: dA, 1,N⁶-ethenodeoxyadenosine; dC, 3,N⁴-ethenodeoxycytidine; LPO, lipid peroxidation; M₁dG, 3-(2-deoxy-D-erythro-pentofuranosyl)pyrimido[1,2-]purin-10(3H)-one; WBC, white blood cell Received September 19, 2008; revised November 13, 2008; accepted November 18, 2008.     

Mutagenicity of nitrosated {alpha}-amino acid derivatives N-acetyl-N' -nitrosotryptophan and its methyl ester in bacteria

DL-N-acetyl-N' -nitrosotryptophan (I) and its methyl ester (II),readily formed under mild conditions by the reaction of nitritewith N-acetyltryptophan or its methyl ester, are model compoundsfor the study of the nitrosation of -amino acid side chains,considered relevant to the possible role of nitrosation of peptidesand proteins in the aetiology of gastrointestinal cancer. Bothcompounds were assayed for mutagenicity in a series of Escherichiacoli WP2 strains (trp^–trp⁺) and in several strains ofSalmonella typhimurium (his^–his⁺), in the presence andabsence of a post-mitochondrial supernatant (S9) from liversof rats treated with Aroclor 1254. Compound I was mutagenicto the following E.coli strains: WP2; WP2uvrA; WP2pKM101; WP2-uvrApKM101;and to S. typhimurium TA 1535, TA 98 and TA 100. Compound IIwas consistently less mutagenic than compound I to the E.colistrains, inactive in S.typhimurium TA 98 and TA 100, but moreactive than I in TA 1535. Neither compound was detectably mutagenicto E.coli WP2 lexA. Addition of S9 did not enhance the mutagenicityof either compound, and in some cases reduced the mutageniceffect. N-Acetyltryptophan alone was not mutagenic to any ofthe E.coli strains tested, and nitrite alone (at pH 7.1) wasvery feebly mutagenic at doses where molar equivalents of compoundI were markedly active. The rate of decay of compound I at pH5.9 was closely paralleled by decay of its mutagenicity. Thesedata and the pattern of cytotoxicity and mutagenicity in severalDNA-repair mutants of E.coli suggest that both compounds reactwith DNA to form excisable DNA-adducts which cause mutationby error-prone repair.     

Intercellular communication in primary cultures of putative preneoplastic and 'normal' hepatocytes

Gap junctional intercellular communication (IC) was studiedin -glutamyltranspeptidase (-GT)-positive and -negative hepatocytesisolated from carcinogen-treated rats. Putative preneoplastic-GT-positive hepatocytes were visualized in monolayer culturesby indirect immunofluorescence using anti--GT-antibodies. ICwas evaluated by studying dye coupling of the cells. -GT-positivehepatocytes showed a significantly lower dye coupling than did-GT-negative liver cells. Spread of the dye Lucifer Yellow CHto neighboring cells was decreased further by the tumor-promotingchemical phenobarbital in both cell types in vitro. Also treatmentin vivo with the barbiturate significantly reduced dye couplingof hepatocytes. The findings suggest that as a result of theirdecreased ability to communicate, preneoplastic hepatocytesmay escape from growth control and differentiation signals givenout by surrounding ‘normal’ cells.     

TGF-{alpha} and EGF-receptor mRNAs in human oral cancer

Transforming growth factor alpha (TGF-) and epidermal growthfactor receptor (EGFR) have been shown to be present in mostsquamous cell carcinomas. Using the Syrian hamster oral cancermodel, we have recently demonstrated the consistent presenceof TGF- and EGFR mRNAs in chemically transformed hamster oralkeratmocytes. We now present evidence that in human oral cancer(in vivo and in vitro), TGF- and EGFR mRNAs can also be consistentlydetected. No TGF- mRNA can be detected in normal human oralepithelium by Northern blot analysis. These findings reinforcethe use of the hamster cheek pouch as an experimental modelfor the study of oral cancer development, at least in referenceto the possible participation of TGF- in the malignant transformationprocess.     

Effects of separate and combined treatments with gamma radiation and diethylnitrosamine in neonatal rats on the induction of altered hepatocyte foci and hepatic tumors

To characterize the effects of combined treatments with gammaradiation and diethylnitrosamine (DEN) on the induction of histochemicallydetectable altered hepatocyte foci and hepatic tumors, we assessedthe yields of these lesions in the livers of 150-day-old ratsthat had been treated neonatally with a single dose of gammaradiation (75 red, whole body) and i.p.-injected DEN (0.15 µmol/gbody wt), either separately or in combination. The combinedtreatments involved the administration of the two stimuli inboth possible sequences, with the interval between treatmentsset at 1 h. The focus population was examined for two histochemicalmarkers (elevated gamma-glutamyl transpeptidase [GGT(+)J andiron exclusion [FE(–)], giving rise to three detectablefocus phenotypes, i.e. GGT(+) foci, FE(–) foci, and GGT(+),FE(–) foci. Frequencies of the three phenotypes were quantitatedthrough the use of serial frozen sectioning techniques and computer-assistedimage analysis. GGT(+) focus induction was synergistkally enhancedby the combined treatment irrespective of the order in whichthe two stimuli were administered; the remaining two phenotypesdid not show such enhancement. The magnitude of the GGT(+) focusresponse was significantly greater when the treatment sequencewas gamma DEN as opposed to DEN gamma. Tumor yields in ratsreceiving combined gamma-DEN treatment were similar to thosein rats receiving the DEN alone, irrespective of the gamma-DENtreatment sequence. These results suggest that (i) phenotypicallydistinguishable lesions, including foci with different histochemicalmarker patterns and tumors, originate from specific types ofdamage at different genetic loci and are developmentally independent;and (ii) the expression of the GGT(+) marker per se in alteredhepatocyte foci is not a reliable index of incipient hepaticneoplasia.     

Immunohistochemical differentiation of {gamma}-glutamyltranspeptidase in focal lesions and in zone I of rat liver after treatment with chemical carcinogens

-Glutamyltranspeptidase (-GT) is known to be increased in putativepre-neoplastic foci but also in the periportal zone I of ratliver under a variety of circumstances not directly relatedto carcinogenesis. To be able to distinguish between these twoinstances -GT was studied by enzyme determination in micro-dissectionsobtained from the two locations and by both histochemical andimmunohistochemical staining in serial sections. Altered hepaticfoci and alterations in zone I were produced in three modelsof hepatocarcinogenesis: (i) initiation by N-nitrosomorpholineand tumor promotion by phenobarbital, (ii) continuous administrationof 2-acetyl-aminofluorene and (iii) continuous administrationof meta-pyrikne hydrochloride. In micro-dissections-GT activitywas similarly increased in focal lesions and in zone I afterfeeding methapyrilene. Histochemically detectable -GT, stainedaccording to Rutenburg et al. (23), was observed both in zoneI and in focal lesions. Focal lesions were also ATPase negativeand UDP-glucuronyhransferase positive in all three models. -GTin focal lesions could be selectively detected by immunohistochemicalstaining using antibodies to the rat kidney enzyme and an indirectperoxidase reaction. These findings suggest immunochemical differencesbetween -GT in focal lesions and in zone I.     

Isolation of {gamma}-glutamyl transpeptidase positive hepatocytes during the early stages of hepatocarcinogenesis in the rat

-Glutamyl transpeptidase (GT) positive hepatocytes were isolatedfrom F344 male rats fed 2-acetylaminofluorene. The isolationprocedure is rapid and highly selective for cells exhibitingGT on their surface. Suspensions of liver cells obtained fromperfusion in situ with a collagenase solution were incubatedon Petri dishes coated with affinity purified rabbit anti-GTantibody. GT-positive cells bound to the dish within fifteenminutes and could be recovered as viable cells. Approximately15% of the GT-positive cells are isolated using this procedure.Novikoff hepatoma cells, a GT-positive cell line, were usedto define the parameters of the assay. The binding was bothtime and temperature dependent. Binding of cells to the anti-GTantibody coated dishes was 50% inhibited by 2.8 mM sodium azideand 86% inhibited by 4.6 mM.     

The influence of cytochrome P-450 induction on the disposition of carcinogenic benzo[a]pyrene 7, 8-dihydrodiol 9, 10-epoxide in isolated cells

The disposition of the carcinogenic (+)-7ß, 8-dihydroxy-9,10-epoxy-7, 8, 9, 10-tetrahydrobenzo[a]pyrene [(+)-anti-BPDE]has been studied in isolated hepatocytes obtained from 3-methylcholanthrene-pretreatedrats. In these cells different routes are acting in concertand contribute to diol-epoxide elimination. Conjugation of (+)-anti-BPDEwith glutathione (GSH) and cytochrome P-450c-mediated metabolismof the diol-epoxide to 1- and 3-hydroxy-anti-BPDE (triol-epoxides)appears to be equally important. The reactive triol-epoxidesundergo a number of secondary reactions, including covalentbinding to cellular constituents, e.g. protein and GSH, andhydrolysis to pentahydroxyderivatives. The effective intracellularlifetime of (+)-anti-BPDE is 1 min and comparable to that previouslyobserved in hepatocytes obtained from uninduced animals.     

Inhibitory effect of {alpha}-tocopherol on the formation of nitrosomorpholine in mice treated with morpholine and exposed to nitrogen dioxide

The possibility that -tocopherol (vitamin E) inhibits the formationof nitrosomorpholine (NMOR) in vivo was investigated in miceorally pretreated with -tocopherol (2.5–100 mg/kg bodywt) once daily for 6 days. Twenty-four hours later, the animalswere injected i.p. with 2 mg of morpholine (MOR) per animalfollowed by exposure txo 47 p.p.m. of NO₂ for 2 h. Under theseconditions, low oral doses of -tocopherol (2.5–5 mg/kgbody wt) significantly decreased NMOR formation in vivo. Astotal body -tocopherol levels increased, in vivo NMOR formationdecreased, and a maximal 50–70% inhibition of NMOR formationoccurred at -tocopherol levels of 5 µg/g body wt. Additionalresults showed that NMOR was rapidly eliminated in mice, sothat studies which measure the levels of NMOR found in animalstreated with MOR and then exposed to NO₂ may underestimate theamount of NMOR that is actually formed. Finally, oral pretreatmentof up to 100 mg of -tocopherol/kg body wt had no effect on NMORelimination.     

Induction of hepatic glutathione S-transferase activity by butylated hydroxyanisole and conjugation of benzo[a]pyrene diol-epoxide

Dietary administration of 2(3)-tert-butyl-4-hydroxyanisole (BHA)to mice caused an increase in the hepatic soluble glutathioneS-transferase activity towards (±)-7ß, 8-dihydroxy-9,10-oxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE) of 5-fold whereasthat towards 1-chloro-2,4-dinitrobenzene (CDNB) was increasedby 14-fold. Whereas with either substrate the catalytic capacityof the enzyme was elevated by BHA treatment, there was littleeffect on the K_m for CDNB but an increase in the K_m for BPDEas substrates. The results thus suggest that BHA-induced GSHS-transferase activity may be of limited importance for protectionfrom certain reactive intermediates of polycyclic aromatic hydrocarbons.