The effect of neighbouring bases on G-specific DNA cleavage mediated by treatment with the anti-diol epoxide of benzo(a)pyrene in vitro

Three 5' -end-labelled double-stranded linear DNA fragmentsof defined sequence were treated with r-7, t-8-dihydroxy-t-9,10-oxy-7, 8, 9, 10-tetrahydrobenzo[a]pyrene (anti-BPDE). TheDNA samples were then examined by gel electrophoresis both beforeand after denaturation and treatment with alkali. The extentof modification of deoxyguanosine (dG) residues was estimatedfrom changes in electrophoretic mobility: at saturation <25% of the dG residues appeared to be modified by reaction withanti-BPDE. TTie determination of the sites of G-specific strandcleavage in a total of 0.5 kbp of DNA by sequencing gel electrophoresisshowed that scission at dG residues is sequence specific andthat whilst, for example, cleavage occurred at the central dGresidues of all 5' -CGG- 3' (21/21), of all 5' - TGG -3' (14/14),of all 5' - TGT -3' (7/7) and of all 5' -CGT-3' (5/5) sequencesexamined, it did not occur in any of the 5' -GGA-3' (0/12) or5' -GGC-3' (0/15) sequences and only occurred rarely in the5' -GGG-3' (1/48) and 5' -GGT-3' (2/11) sequences. No cleavagewas found at internal dG residues within poly(dG)9 or poly(dG)18sequences. The data may permit prediction of the sites of strandscission in DNA molecules of known sequence that have been modifiedby diol-epoxides of polycyclic hydrocarbons.     

Sequence specificity of guanine alkylation and repair

The sequence selectivity of methylation at the O⁶ and N7 positionof guanine by N-methyl-N'-nitrosourea (MNU) and the rate ofremoval of O⁶-methylguanine by O⁶ alkylguanine-DNA alkyltransferase(AGT) was determined using dodecadeoxynucleotides of definedstructure. The extent of guanine adduct formed in self-complementarydodecamers, 5'- TATACGCGTATA-3', 5'-TATACCGGTATA-3' and 5'-TATAGGCCTATA-3'AND 5'-TATAGGCCTATA-3', after methylation with [³H]MNU in arepresentative experiment were, respectively, 10, 19 and 30pmol O⁶-methylguanine/µmol guanine and 97, 189 and 217pmol N7-methylguanine/µmol guanine. The O⁶-methylguanine/N7-molmethylguanineratio remained relatively constant for each dodecamer. A directcomparison between the methylation at guanine with adenine orthymine as the 5'-flanking base was made with two dodecamers,5'-TATACATGTATA-3' and 5'-TATACTAGTATA-3'. When the guanineresidue was preceded 5' by an adenine, the level of O⁶ and N7-alkylationwas, respectively, 2.1-fold and 1.5-fold greater than when guaninewas preceded 5' by a thymine. These date are consistent witha regioselective mechanism for alkylnitrosourea alkylation ofguanine. The methylated dodecamer, 5'-TATACGCGTATA-3' was repairedfaster than 5'-TATACCGGTATA-3' by HT29 extract containing AGTwith a loss in 10min of 0.052 pmol and 0.025 pmol O⁶-methylguanine,respectively. Dodecamers of the structure 5'-dCGCGAATTCm⁶GCG-3'and 5'-dCGCCAATTGm⁶GCG-3' were labeled at the 5' end with ³²Pby the reaction with polynucleotide Kinase and after incubationwith AGT, the methylated and demethylated dodecamers were separatedby reversed-phase HPLC. The amount of demethylated product formedwas greater for the dodecamer containing cytosine as the 5'-flankingbase to O⁶-methylguanine compared to guanine in that same position.A higher extent of alkylation by MNU and a slower rate of repairby AGT for sites in which a guanine or modified guanine is precededby a purine rather than a pyrimidine may explain, at least inpart, mutational hot spots.     

Use of fluorescently tagged DNA and an automated DNA sequencer for the comparison of the sequence selectivity of SN1 and SN2 alkylating agents

This paper describes the application of the novel non-radioactivetechnique for studying the sequence selectivity of selectedalkylating agents. N-Nitroso-N-methylurea (MNU) and N-methyl-N'-nitro-nitrosoguanidine(MNNG) were chosen from the S^N1 group of alkylating agents.Dimethyl sulphate (DMS) was used to represent alkylation profileproduced by the S_N2 compounds. Results of S_N1 compounds indicatedthat in a run (G)₃ the latter two Gs are more susceptible toalkylation than the most 5' G. Moreover, in a GG sequence the3' G seems to be more alkylated. This effect is more evidentwhen the GG site was preceded by a 5' pyrimidine. These findingssuggest that a regio-selective mechanism, rather than the formationof diazonium ions, accounts for DNA alkylation by S_Nl compounds.On the other hand, DMS showed preferential alkylation of the5' end in a (G)₃ run. However, at GG sequences no clear preferredsite of alkylation could be distinguished. Lack of specificityof S_N2 compound would seem to suggest that other factors aswell as the primary DNA structure may play a role in determiningthe extent of alkylation at a certain site.     

Regiospecificity in the metabolism of the homologous cyclic nitrosamines, N' -nitrosonornicotine and N' -nitrosoanabasine

We compared the metabolism in the F-344 rat of the moderatelypotent esophageal carcinogen N' -nitrosonornicotine (NNN, 2'-(3-pyridyl)-N-nitrosopyrrolidine) and its weakly active homologueN'-nitrosoanabasine (NAB, 2' -(3-pyridyl)-N-nitrosopiperldine).Urine was the major pathway of excretion for both nitrosamines.The major urinary metabolites of dl-NNN resulted from 2' -hydroxylation(8.1% of the dose), 5' -hydroxylation (37.6%), and pyridineN-oxidation (10.8%). The percentages of the dose of the correspondingmetabolites of dl-NAB were: 2'-hydroxylation (not detected),6'-hydroxylation (9.8%), pyridine-N-oxidation (30.0%). Similarresults were obtained when the urinary metabolites of l-NNNand l-NAB were compared. In 48 h cultures of rat esophagus,the major metabolites of [2' -¹⁴C]dl-NNN resulted from 2' -hydroxylation(47%) and to a lesser extent from 5'-hydroxylation (15%). Incontrast the major metabolite of [2'-¹⁴C]dl-NAB resulted from6'-hydroxylation (35%) with lesser amounts from 2'-hydroxylation(8%). 6'-Hydroxylation of [2'-¹⁴C]dl-NAB also exceeded 2' -hydroxylationin cultures of 3,6,12 or 24 h duration. Pyridine-N-oxidationwas not observed in the esophagus for either nitrosamine. Theseresults demonstrate a high degree of regiospecificity in themetabolism of these structurally related nitrosamines. Amongthe identified urinary metabolites the ratio of -hydroxylationto N-oxidation was 4.2 for NNN and 0.3 for NAB. Among the 48h esophageal metabolites the ratio of 2'-hydroxylation to 5'- or 6' -hydroxylation was 3.1 for NNN and 0.2 for NAB. Theresults also suggest a basis for the weak carcinogenicity ofNAB: facile excretion as its pyridine-N-oxide and detoxificationin the esophagus by 6' -hydroxylation.     

Mutation specificity of 8-methoxypsoralen plus two doses of UVA irradiation in the hprt gene in diploid human fibroblasts

To investigate which specific kinds of base changes are inducedby psoralen adducts in the genomic DNA of diploid human fibroblasts,cells were exposed to 8-methoxypsoralen (8-MOP) at 2 –12µM followed by one dose of UVA (365 nm) irradiation (PUVA-Itreatment) or two doses of UVA (PUVA-II treatment). While PUVA-Itreatment produced little effect on the induction of cytotoxicity,PUVA-II treatment significantly reduced the fibroblasts' colony-formingability and resulted in about 10-fold increases in mutationfrequency at the Do dose. Mutations in the hypoxanthine (guanine)phosphoribosyltransferase (hprt) gene of 36 independent PUVA-IImutants were characterized by direct sequencing of cDNA amplifiedby the polymerase chain reaction (PCR). Seventeen mutants containedsingle base substitutions and the other 19 mutants either lackedone or more exons, or had deleted or gained nucleotides in theexon boundaries in their cDNA. The intron-exon boundaries of10 of these 19 putative splicing mutants were further characterizedby direct sequencing of the PCR-amplified hprt gene. The resultsshowed that nine contained single base substitutions at theconsensus splicing donor and acceptor sites. One splicing mutantpossessed two base substitutions located at exon 8, whereasits splicing sites were intact. Most of the base substitutionsoccurred at TA base pairs (24/29). The majority of TA changesoccurred at thymine of 5'TA and 5'ATA on the non-transcribedstrand. Four of the five GC base substitutions were locatedat guanines of 5'TG sites adjacent 3' to AT or TA sequences.In addition, the occurrence of a specific type of mutation washighly correlated to the 5' flanking bases of TA sites. Themutagenesis of 13 of the 16 mutational events at 5'TA siteson the non-transcribed strand can be explained by the preferentialincisions of the photoadducts on the transcribed strand followedby misalignment realignment during translesion repair synthesisof the bulky lesions on the non-transcribed strand.     

Identification of N2-(deoxyguanosin-8-yl)-2-amino-3,8-dimethylimidazo[4,5-f)quinoxaline 3',5'-diphosphate, a major DNA adduct, detected by nuclease P1 modification of the 32P-postlabeling method, in the liver of rats fed MeIQx

The carcinogenic heterocyclic amine 2-amino-3,8-dimethyl-imidazo[4,5-f]quinoxaline(MeIQx) is widely distributed in cooked foods. The nucleaseP1 method increased the sensitivity of the standard ³²P-postlabelinganalysis about 1000-fold for detection of MeIQx-DNA adducts.The recovery of MeIQx-DNA adducts by the nuclease P1 methodwas determined to be about 50% using liver DNA of a rat treatedwith [¹⁴C]MeIQx intragastrically. By the nuclease P1 methodfive adducts were detected in the liver DNA of rats fed MeIQxand two of them, including the most abundant one, were identifiedas MeIQx-deoxyguanosine adducts by comparison with the adductsformed in in vitro reactions of N-acetoxy-2-amino-3,8-dimethylimidazo[4,5-f)quinoxalinewith the four 2'-deoxyribonucleotides. The most abundant adductin vivo was identified as N²-(deoxyguanosin-8-yl)-MeIQx 3',5'-diphosphate(3',5'-pdGp-C8-MeIQx). MeIQx-DNA adduct levels in human tissuescould be determined by the nuclease P1 modification of the ³²P-postlabelingmethod in combination with HPLC, and thus provide informationon the roles of MeIQx in human carcinogenesis.     

Mutational specificity of N-methyl-N-nitrosourea in the lacI gene of Escherichia coli

The mutational Specificity of the carcinogenic alkylating agentN-methly-N-nitrosourea (MNU) was investigated through the DNAsequence characterisation of 104 lacI^-d mutations induced bythis agent in the loci gene of Escherichia coli. The vast majority(95%) of MNU-induced mutations were G:C A:T transitions. Ananalysis of neighbouring base sequence revealed that this transitionwas almost 10 times more likely to occur if the mutated guanineresidue was preceded (5') by a purine (R) rather than a pyrimidine(Y); apart from this 5'-R-G-3' influence no other site specificityof mutation was observed. This 5' flanking base influence onmutability has also been observed in this system with othermechanistically similar alkylating agents.     

Cisplatin and protracted venous infusion 5-fluorouracil (CF) -- good symptom relief with low toxicity in advanced pancreatic carcinoma

Purpose: The combination of protracted venous infusion (PVI)5-FU with moderate dose cisplatin was evaluated in patientswith advanced pancreatic cancer Patients and methods: Sixty-three patients with locally advancedor metastatic disease were treated with cisplatin (60 mg/m²every 21 days) and PVI 5-FU (300 mg/m²/day) for a maximum of24 weeks. All patients had histologkally/ cytologically confirmedtumour. Radiological response was assessed by CT scanning andtoxicity, performance status and symptomatic response were assessed3 weekly Results: The objective response rate was 16% with two radiologicalcomplete responses. The median survival was 7.6 months witha 1-year survival of 33% and a median progression-free survivalof 6.6 months. Patients who had local disease only had a mediansurvival of 14.8 months with a 1-year survival of 52%. Thirty-fourpercent of patients had an improvement in performance statuson treatment and specific symptom response rates were weightloss 71%, dysphagia 100%, nausea and vomiting 70%, pain 60%,anorexia 50% and reflux 81%. Chemotherapy was well toleratedwith grade 3 or 4 toxicity being nausea/vomiting 5%, diarrhoea7%, infection 4%, stomatitis 2%, plantar palmar syndrome 2%,anaemia 14%, leucopenia 5%, neutro-penia 10% and thrombocytopenia8% Conclusions: The CF regimen provides good symptomatic palliationwith low toxicity in patients with advanced pancreatic cancer pancreatic cancer, 5-fluorouracil, cisplatin     

Detection of deoxyadenosine-4-aminobiphenyl adduct in DNA of human uroepithelial cells treated with N-hydroxy-4-aminobiphenyl following nuclease P1 enrichment and 32P-postlabeling analysis

To characterize the DNA adducts in human uroepithelial cells(HUC) exposed to 4-aminobiphenyl and its proximate N-hydroxymetabolites, we used ³²P-analyses following butanol extractionof the DNA hydrolysates. Using this method, we identified N-(deoxyguanosin-3',5'-bisphospho-8-yl)-4-aminobiphenyl (pdGp-ABP) as a major adductand N-(deoxyguanosin-3', 5'-bisphospho-8-yl)-4- aminobiphenyl(pdAp-ABP) as a minor adduct in an immortalized non-tumorigeniccell line of HUC following exposure to N-hydroxy-4-aminobiphenyl(N-OH-ABP). Towards characterization of pdAp-ABP, we postlabeledthe synthetic N-(deoxyguanosin-3', 5'-bisphospho-8-yl)-4-aminobi-phenyl (dAp-ABP) adduct to generate pdAp-ABP and determinedits chromatographic (TLC and HPLC) proper ties and sensitivityto nuclease P1 digestion. In contrast to pdGp-ABP, which wascleaved to the corresponding 5' monophosphate by nuclease P1,the pdAp-ABP adduct was unaffected when incubated with nucleaseP1 under similar conditions. To test whether nuclease P1 digestioncould be adopted for enrichment of the dAp-ABP adduct in HUCsamples, postlabeling analyses were carried out after but anolextraction following nuclease P1 digestion of the DNA hydrolysate.Under these conditions, the pdAp-ABP adduct was detected inDNA from HUC E7 cells treated with N OH-ABP and in calf thymusDNA reacted with N-OH ABP under acidic (pH 5.0) conditions.These data indicate that pdGp-ABP and pdAp-ABP adducts are generatedin HUC E7 on treatment with N-OH-ABP and that nuclease P1 enrichmentmay provide a method for qualitative and quantitative analysesof the pdAp-ABP adduct in DNA.     

Effect of 5-methylcytosine as a neighboring base on methylation of DNA guanine by N-methyl-N-nitrosourea

Effects on N-methyl-N-nitrosourea (MNU) mediated methylationof the N7 position of guanine were compared in defined sequencesof DNA containing cytosine or 5-methylcytosine (5mC) using aMaxam-Gilbert sequencing technique. Cytosine methylation in5'-CpG-3' pairs within a subcloned fragment of the 5' regionof the human HPRT gene was generated with SssI methylase andS-adenosylmethionine. Cytosine methylation was demonstratedby both the inhibition of DNA restriction by methylation sensitiveendonucleases and the lack of cleavage at 5-methylcytosinesby hydrazine. MNU-dependent methylation of the N7 position ofguanine was inhibited up to 18% when 5mC was a 5' neighboringbase to guanine and was inhibited up to 36% in an alternatingCpG region in which both 5' and 3' neighboring bases of guaninewere enzymatically altered to 5mC. It can be concluded that5-methylcytosine has discernible effects on MNU methylationof the N7 position of specific guanine bases in DNA.     

Altered fidelity of a nucleic acid modifying enzyme, T4 polynucleotide kinase, by safrole-induced DNA damage

Mouse liver DNA adducted with metabolites of the spice constituentsafrole (1-allyl-3, 4-methylenedioxybenzene), when analyzedvia the bisphosphate version of the ³²P-postlabeling assay,exhibits two major adducts, which had been previously identifiedas N²-(trans-isosafrol-3' -yl)2' -deoxyguanosine 3', 5'-bisphosphate(adduct 1) and N²(safrol-1'-yl)2'-deoxyguanosine 3', 5'-bisphosphate(adduct 2). However, analysis of the same DNA preparation bythe dinucleotide/monophosphate version of the assay gave twoadditional spots on PEI-ceihilose TLC whose nature was clarifiedin the present study. Several enzymes (T4 polynucleotide kinase,nuclease P1, venom phosphodiesterase and spleen phosphodiesterase)were utilized to hydrolyze these compounds, and the productsco-chromatographed on PEI-cellulose thin layers with radiolabeledand non-radioactive nucleotides of known structure. The additionalspots were found to be adducted dinucleotides carrying ³²P-labelat both the 5'- and 3'-hydroxyls. T4 polynucleotide kinase-catalyzed3'-phosphorylation was highly specific in that only dinucleosidemonophosphate derivatives of adduct 1, with an unmodified purinein the 3'-position, were susceptible to both 5'- and 3'-phosphorylationby the enzyme. Thus, the structures of the two additional ³²P-labeledsafrole derivatives were pX₁pAp and pX₁pGp where X₁ denotesN²-(trans-isosafrol-3'-yI)2'-deoxyguanosine. The official nameof T4 polynucleotide kinase, ATP: 5'-dephosphopolynucleotide5'-phosphotransferase (EC 2.7.1.78 [EC] ), denotes the specific actionof this enzyme as a 5'-phosphokinase. Although the enzyme has3'-phosphatase activity at acidic pH, no 3'-kinase reactionhas been previously reported. Possible implications for chemicalcarcinogenesis of the finding that carcinogen - DNA adductscan specifically alter the fidelity of protein-nucleotide interactionsare discussed.     

Comparison of mutation spectra induced by N-ethyl-N-nitrosourea in the hprt gene of Mer+ and Mer- diploid human fibroblasts

N-Ethyl-N-nitrosourea (ENU) forms several major adducts uponreaction with DNA, of which ethylation at the O⁶ position ofguanine and the O⁴, O² and N³ positions of thymine have beenimplicated to be mutagenic lesions. To investigate what specifickinds of ENU-induced mutations were affected by the repair abilityof O⁶-alkylguanine-DNA alkyltransferase (AGT), we examined themutations in the hypoxanthlne (guanine) phosphoribosyltransferasegene (hprt) in 87 independent mutants derived from ENU-treatedAGT proficient (Mer⁺) or deficient (Men^–) diploid humanfibroblasts. Of the characterized mutations, 97% were singlebase substitutions. The major difference in the mutation spectrawas that the frequency of G·C to A·T transitionswas significantly higher in Mer^– mutants (16/38) thanin Mer mutants (4/33). The results indicate that AGT removesO⁶-ethylguanine, thus protecting human cells from parts of thecytotoxic and mutagenic effects of ENU. A high frequency ofT· to A·T transversions induced by ENU was observedin both Mer⁺ (52%) and Mer^– (34%) mutants. This type ofmutation was less frequently observed (10%) in N-methyl N'-nitro-N-nitrosoguanidine(MNNG)-induced mutants derived from the same Mer⁺ cells in ourprevious report (J. Mol. Biol., 221, 421, 1991). Comparisonof alkylating lesions formed by MNNG and ENU indicates thatO² and N³-ethylthymine are potent mutational adducts for T toA transversions. The occurrence of ENU induced T·A basepair transverslons showed a strong strand bias; 35/37 were locatedon the non-transcribed strand, assuming thymine is the mutageniclesion. The result suggests a difference in repair capacityof ethyithymine on the two strands. In addition, this type ofmutation preferentially occurred at 5'-Pu-T sequences.     

Optimization of 32P-postlabelling assays for the quantitation of O6-mettyl and N7-methyldeoxyguanosine-3' -monophosphates in human DNA

The 3' and 5'-monophosphates of O⁶-methyldeoxyguanosine andN7-methyldeoxyguanoslne were chemically synthesized. Using thesestandards with deoxyguanoslne-3'-monophosphate (dGp) as an internalstandard, conditions were optimized to quantify O⁶ -methyldeoxyguanosine-3'-monophosphate(O⁶ and N7-methyldeoxyguanosine-3'-monophosphate (N7-MedGp)by ³²P-postlabelling Under optimal conditions, the labellingefficiencies of O⁶ -MedGp and N7-MedGp were respectively     

Phase II trial of 5-fluorouracil, leucovorin and cisplatin for treatment of advanced pancreatic adenocarcinoma

Background. Advanced pancreatic adenocarcinoma is a rapidlyfatal disease for which an active chemotherapy regimen is sought.Here we report the outcome of a phase II trial to assess thetoxicity and efficacy of a combination of 5-flu-orouracil (5-FU),leucovorin and cisplatin (CDDP). Methods. A regimen combining leucovorin (200 mg/m²/d x 5d),5-FU (375 mg/m²/d x 5d in a 2-hour infusion) and CDDP (15 mg/m²/dx 5d) was given to 52 patients with histologically-proven, previouslyuntreated, locally advanced (n = 13) and/or metastatic (n =39) pancreatic adenocarcinoma. Results. Of 48 patients evaluable for response, 10 achievedpartial responses, for an overall response rate of 21% (95%CI 9.5%–32.5%), and a palliative effect was observed in52%. The median survival was 9.5 months (18 months for locally-advancedand 5 months for metastatic disease) with a 1-year survivalof 34.6% and a median progression-free survival of 4.5 months.Chemotherapy was well tolerated with grades 3 or 4 nausea/vomitingin 12%, diarrhea in 6%, anaemia in 17%, neutropenia in 12%,and thrombocytopenia in 10%. Eleven patients (21%) had Grade2 peripheral neuropathy. Conclusion. The combination of leucovorin, 5-FU and CDDP seemsto be an effective palliative treatment, with moderate toxiceffects, in advanced pancreatic adenocarcinoma. advanced pancreatic adenocarcinoma, cisplatin, 5-fluorouracil, leucovorin     

Identification of N-(deoxyguanosin-8-yl)-2-amino-3,4-dimethylimidazo[4,5-f]quinoline (dG-C8-MeIQ) as a major adduct formed by MeIQ with nucleotides in vitro and with DNA in vivo

The N-hydroxylamine of a carcinogenic heterocyclic amine, 2-amino-3,4-dimethylimidazo[4,5-f]quinoline(MeIQ), was reacted with four 2'-deoxynucleoside 3'-monophosphatesafter O-acetylation. ³²P-Postlabeing analysis demonstrated thatthe adduct was formed with only the guanine nucleotide, andthe structure of the compound in the obtained adduct spot wasdetermined to be N-(deoxyguanosin-8-yl)-MeIQ 3',5'-diphosphate(3',5'-pdGp-C8-MeIQ). DNA samples from livers of mice fed MelQwere also ³²P labeled under standard conditions and additionallytreated with nuclease P1 and phosphodiesterase I. A single adductspot was obtained and the structure of the adduct was identifiedas 5'-pdG-C8-MeIQ. Thus, MelQ binds at the C-8 position of guaninein vitro and in vivo, like other heterocyclic amines.     

A unique p53 intragenic deletion flanked by short direct repeats results in loss ofmRNA expression in a human esophageal carcinoma

A 45 base pair (bp) intragenic deletion of the p53 gene froma human esophageal cancer was analyzed in detail. This deletioncontained all RNA splice consensus sequences at the 3' end ofintron 7, including the RNA splice branch point, the pyrimidine-richregion and the 3' splice acceptor site. Northern blotting revealeda total lack of p53 mRNA expression in this tumor. Short directrepeats (TACTG) were found at the 5' and 3' breakpoints of thedeletion and it removed one complete repeat as well as the entireregion between the repeats. These results suggest that a 'slippedmispairing' mechanism occurring during DNA replication may generatep53 intragenic deletion in human esophageal cancer, leadingto abolished p53 mRNA expression.     

O6-substituted-2'-deoxyguanosine-3'-phosphate adducts detected by 32P past-labeling of styrene oxide treated DNA

³²P post-labeling of DNA reacted with styrene oxide resultedin the detection of six adducts. In order to determine whichof these corresponded to modification at the O⁶ position ofguanine, O⁶-substituted styrene oxide-deoxyguanosine–-3'–monophosphatederivatives were synthesized. The two synthetic isomers werepurified by HPLC and the structures were confirmed by mass spectrometryand ¹H NMR. ³²P post-labeling and co-chromatography with theDNA-styrene–7, 8–oxide reaction products resultedin the assignment of adduct number 4 as O⁶hydroxy-2-phenylethyl)-2'-deoxyguanosine-3',5'-bisphosphate and adduct number 5 as O⁶-(2-hydroxy-1-phenylethyl)-2'-deoxyguanosine-3',5'-bishosphate.     

DNA adduct formation in Salmonella typhimurium, cultured liver cells and in Fischer 344 rats treated with o-tolyl phosphates and their metabolites

2-Phenoxy-4H-1,3,2-benzodioxaphosphorin 2-oxide is an electrophilicand a neurotoxic metabolite of o-tolyl phosphates. In a previouspaper we reported that 2-phenoxy-4H-1,3,2-benzodioxaphosphorin2-oxide is mutagenic in Salmonella typhimurium TA100 and formsDNA adducts in incubations with nucleotides, nucleosides andisolated DNA. In the present study we compare DNA adduct formationusing ³²P-post-labelling assays in 2-phenoxy-4H-1,3,2-benzodioxaphosphorin2-oxide-treated bacteria (S.typhimurium TA100) and hepatomacells with DNA adducts formed in liver, kidney, lung and heartof tri-o-tolyl phosphate-exposed Fischer 344 male rats. In bothbacteria and hepatoma cells two DNA adducts could be detectedafter treatment with 2-phenoxy-4H-1,3,2-benzodioxaphosphorin2-oxide. The minor adduct co-chromatographed with syntheticN³-(o-hydroxy-benzyl)deoxyuridine 3' monophosphate after postlabelling.The major DNA adduct was a cytidine adduct, most likely N³-(o-hydroxybenzyl)deoxycytidine3' monophosphate. Male Fischer 344 rats were treated orallyfor 10 days with tri-o-tolyl phosphate (50 mg/kg/day) and DNAwas isolated from liver, kidney, lung, heart, brain and testes1,4,7 and 28 days after giving the last dose. Analysis by ³²P-postlabellingrevealed that two adducts were present in the DNA isolated fromliver, kidney, lung and heart on the first day after givingthe last dose; DNA adducts were not detected in the brain andtestes. The adduct pattern after in vivo treatment with tri-o-tolylphosphate was identical with that found in bacteria and hepatomacells treated with 2-phenoxy-4H-1,3,2-benzo-dioxaphosphorin2-oxide, the major adduct being N³-(o-hydroxybenzyl)deoxycytidine3' monophosphate and the minor N³-(o-hydroxybenzyl)deoxyuridine3' monophosphate. Both DNA adducts persisted in the lungs forthe entire observation period, whereas in the kidney only thecytidine adduct could be detected 28 days after the last doseof tri-o-tolyl phosphate. In liver and heart the adducts weredetectable only on the first day after completion of the treatment.The results indicate that in addition to the well establishedneurotoxicity, some o-tolyl phosphates may have a carcinogenicpotential.     

Detection by 32P-postlabeling of thymidine glycol in {gamma}-irradiated DNA

The ³²P-postlabeling method has been adapted for the analysisof thymidine-cis-glycol-3'-phosphate (cis-dTGp, cis-5,6-dihydroxy-5,6-dihydrothymidine-3'-phosphate).Cis-dTGp was isolated and purified from normal nucleotides byphenylboronate affinity chromatography and phosphorylated byT4 polynucleotide kinase in presence of 1 mM BeCl₂ at pH 7.5.These modifications of the postlabeling method resulted in a5'-phosphorylation of dTGp with a labeling efficiency of upto 20% whereas the natural nucleotides were almost completelydephosphorylated at the 3' position under these conditions.The reaction products, containing radio-labeled thymidine-cis-glycol-3',5'-bis-[5'-³²P]phosphate (cis-^*pdTGp), were separated by two-dimensionalanion-exchange TLC on polyethyleneimine cellulose sheets. Boricacid was added in the second dimension in order to selectivelyretard cis-glycols. The method was applied to -irradiated nucleotidesand calf thymus DNA. In the nucleotide mixture, 330–99000 thymine glycol (TG) moieties were detected per 10⁶ thymines(T) in a dose range of 14–1000 Gy respectively. In DNA,these values ranged from 400 to 2700 TG/10⁶ T. The data arein good agreement with methods using radiochemical and immunologicaltechniques. Non-irradiated DNA showed a background level of1OTG/10⁶ T. This practical limit of detection was higher thancan be achieved with the postlabeling technique, indicatingthat the present method might be a sensitive alternative fora determination of oxidative DNA damage.     

Specificity of mutations induced by N-methyl-N-nitrosourea in a cDNA of the hprt gene

N-Methyl-N-nitrosourea (MNU) reacts with DNA to produce a varietyof lesions, of which O⁶-methylguanine (O⁶-MeG) has been implicatedin the mutagenic and carcinogenic effects of this agent. Thepresent study aimed to investigate the types and position specificitiesof mutations induced by MNU. Mutational sequence alterationswere determined for 53 independent mutations induced by MNUin a cDNA of the human hprt gene, which is stably integratedinto chromosomes of the mouse cell line VH12. The majority ofthe mutations induced by MNU were base substitutions (85%),mostly GC to AT transitions (41/43), and the remainder (15%)were frameshift or deletion mutations resulting from loss oraddition of a few bases. The transition mutations occurred preferentiallyat the middle G in 5'-purine-G-N-3' sequences in the non-transcribedstrand, and were distributed non-randomly over the coding regionof the gene. Analysis of the results suggests that, when interpretingmutational specificity and distribution, not only the natureof the mutational target sequence, but also the functional domainsof the protein should be considered.