[CANCER RESEARCH 45, 2440-2444, June 1985]

Lack of Miscoding Properties of 7-(2-Oxoethyl)guanine, the Major Vinyl

Chloride-DNA Adduct

Alain Barbin, Reinhold J. Laib1 and Helmut Bartsch2

International Agency for Research on Cancer, Division of Environmental Carcinogenesis, 150 Cours Albert Thomas, F 69372, Lyon Cedex 08, France [A. B., H. B], andInstitut fürArbeitsphysiologie an der UniversitätDortmund, Ardeystrasse 67, D 4600 Dortmund, Federal Republic of Germany [P. J. L]

ABSTRACT

Chloroethylene oxide, an ultimate carcinogenic metabolite ofvinyl chloride, was reacted with poly(deoxyguanylate-deoxycyti-dylate); the nucleic acid base adducts, 7-(2-oxoethyl)guanine andS^-ethenocytosine, were analyzed by reverse-phase high-performance liquid chromatography. Chloroethylene oxide-modifiedpoly(deoxyguanylate-deoxycytidylate) was assayed as templatein a replication fidelity assay with Escherichia coli DNA polym-

erase I, and the newly synthesized product was subjected tonearest-neighbor analysis. Misincorporation rates of deoxyaden-

osine monophosphate and thymidine monophosphate werefound to increase with the level of template modification. About80% of the mispairing events were located opposite minorcytosine lesions. 7-(2-Oxoethyl)guanine, the major adduct iden

tified (>98% of the adducts), did not miscode for either thymineor adenine, failing to support an earlier hypothesis that the cyclichemiacetal form, O6,7-(1'-hydroxyethano)guanine, could, byanalogy with O6-methyl- and O6-ethylguanine, simulate adenine.

Our results indicate that direct miscoding of 7-(2-oxoethyl)~

guanine may contribute only slightly to the induction of mutationsby Chloroethylene oxide or vinyl chloride.

INTRODUCTION

The mutagenic and carcinogenic activity (4, 9, 15) of vinylchloride has been shown to be dependent upon its metabolicactivation into CEO3 (Ref. 3; for a recent review, see Ref. 12).

The carcinogenicity of CEO was demonstrated in mice (27), butno evidence has been obtained to date that chloroacetaldehyde,a rearrangement product (3), is carcinogenic (6, 26, 27).

CEO and chloroacetaldehyde can both bind covalently tonucleic acid bases in vitro to yield 1,N6-ethenoadenine, «C,and

Af.S-ethenoguanine (1,11,16); oxet-G is produced by reactionof CEO with deoxyguanosine (21). In a microsome-mediated

assay, addition of vinyl chloride led to the formation in DMA ofoxet-G, 1./V-ethenoadenine, and «C(14). After rats were exposed for 2 years to vinyl chloride, 1./v^-ethenoadenine and «C

were reported to have been formed in DNA (5); however, following shorter exposures to vinyl chloride, oxet-G was found to be

the major DNA adduct formed in rats (13,14) and mice (19).By analogy to the promutagenic lesion O6-alkylguanine (22), it

was suggested that the genetic and possibly carcinogenic effectsof vinyl chloride might result from the persistence of miscoding

1Part of this work was performed under an IARC Research Training Fellowship

at the International Agency for Research on Cancer, Lyon, France.2To whom requests for reprints should be addressed.3The abbreviations used are: CEO, chtoroethytene oxide; oxet-G, 7-(2-oxo-

ethyOguanine; «C,S.W-ethenocytosine; HPLC, high-performance liquid chromatography; poly(dG-dC), poly(deoxyguanylate-deoxycytidylate); AP, apurinic-apyrimi-

dinic.Received 12/13/83; revised 8/14/84,12/11/84; accepted 2/21/85.

DNA adducts (3, 16, 21, 27). Previous reports (2, 7, 23, 25)indicated a slightly decreased fidelity of replication of syntheticpolynucleotides with Escherichia coli DNA polymerase I, whenthe templates contained 1.W-ethenoadenine or «C.As an exten

sion of these studies, we have now investigated the codingproperties of oxet-G, which has been proposed to miscode forthymine, following the formation of a cyclic hemiacetal at the O6-

position (Ref. 21 ; see Chart 1).

MATERIALS AND METHODS

Treatment of Templates

Poly(dG-dC) (820.»= 8.0, 20.3 AMOunits/mg, from P-L Biochemicals

GmbH, St. Goar, Federal Republic of Germany) was prepared by dissolving 4.35 mg in 14.5 ml 50 mw sodium cacodylate buffer (pH 7.0):0.2M sodium chloride. CEO [purity >99.5%, prepared and stored as described previously (3)] in 1- to 5-/<l aliquots was added under constantstirring at room temperature (21 °C).A pH of 7.0 was maintained carefully

by addition of 2 N NaOH using an autoburet and an automatic titrator.Under these conditions, CEO reacted very rapidly (half-life, about 3 min).

The end of the reaction, after each addition, was indicated by a stablepH; another CEO aliquot was then immediately added and the pH wasstabilized again. This operation was repeated until the desiredCEO:nucleotide phosphorus ratio, i.e., 12, was attained. A 1.8-ml aliquotof the reaction medium was collected and stored at -70°C. Addition to

the remaining poly(dG-dC) solution of CEO was continued as described

above, until the CEO added:nucleotide phosphorus ratio was 45; afurther 1.8-ml aliquot of the reaction medium was then collected andstored at -70°C. These operations were continued for further modifi

cation of the polynucleotide. The 1.8-ml aliquots of the reaction mediumwere thawed and passed through a Sephadex G-50F column (1 x 35.5

cm; Pharmacia Fine Chemicals, Uppsala, Sweden) with water as eluant,in order to remove salts and excess chloroacetaldehyde. Polymer-containing fractions were lyophilized; each batch of treated poly(dG-dC) wasdissolved in 50 HIM Tris-HCI buffer (pH 7.8) and stored at -20°C. The

following CEOmucleotide phosphorus molar ratios were used to modifythe templates: 12, 45, 89, 173, 206, 256, 291, and 318; the treatedtemplates are referred to, respectively, as I, II, III, IV, V, VI, VII, and VIII(Table 1). In some experiments, Templates VI, VII, and VIII were pooled;this pooled template is referred to as Template IX. Untreated polyfdG-dC) [control poly(dG-dC)] was processed in the same way as were theCEO-modified templates.

HPLC Techniques

HPLC separations were performed on a DuPont 850 liquid Chromatograph (DuPont de Nemours S.A., Les Ulis, France); using a Hewlett-Packard 3390 A integrator (Hewlett-Packard, Orsay, France), UV absor-

bance was monitored at 254 nm. Mobile phases were prepared withMilli-Q water (Millipore water purification system); methanol (LiChrosolv,

HPLC grade) and other chemicals (all analytical reagent grade) wereobtained from Merck AG, Darmstadt, Federal Republic of Germany.

Analysis of CEO-treated Poly(dG-dC). Aliquots containing 100 ngCEO-treated poly(dG-dC) were lyophilized and subjected to acid or

CANCER RESEARCH VOL. 45 JUNE 1985

2440

on July 12, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

EFFECTS OF CEO-INDUCED LESIONS ON DNA REPLICATION

HO

(dR) 0---H —N

HIChart 1. Hypothesized miscoding properties of oxet-G. Proposed equilibrium

between oxet-G (I) and the hemiacetal (II) and mispairing scheme of II with thymine(III) (21). dR, deoxyribose.

enzymic hydrolysis. The former was performed in 200 n\ 0.1 N hydrochloric acid (45 min, 75°C),and samples were analyzed by HPLC without

further treatment. For enzymic hydrolysis, samples were dissolved in200 n\ 50 rtiM Tris-HCI (pH 6.75) containing 4 mg (8000 units/ml) DNase

I (EC 3.1.21.1 ; Boehringer Mannheim GmbH, Mannheim, Federal Republic of Germany) and incubated for 3 h at 37°C. Subsequently, 10 //I (10

f¡g,0.015 units with bis-4-nitrophenyl phosphate as substrate) snake

venom phosphodiesterase (EC 3.1.4.1 ; Boehringer Mannheim) and 10 n\(10 ¿»g,0.06 unit) acid phosphatase (EC 3.1.3.2; Boehringer Mannheim)were added, and incubation was continued for a further 15 h. Precipitatedproteins were removed by cent rifuga t¡on,and the supernatant was

analyzed by HPLC.Oxet-G was analyzed following either enzymatic or acid hydrolysis

(14). Enzymatic hydrolysis was used to release 3,W-ethenodeoxycyfidinebecause it is unstable under acidic conditions. W2,3-Ethenoguanine was

analyzed by acid hydrolysis.The nucleosides and bases generated by hydrolysis of CEO-modified

poly(dG-dC) were separated by gradient elution on a 4.6-mm x 25-cmRP-18 Spheri-5 reverse-phase column (Brownlee Labs, Inc., Santa Clara,

CA). Eluant A was 0.025 M ammonium formate (pH 5.5), and Eluant Bwas methanol: water (80:20, v/v). The gradient program was: Step 1,0%Eluant B (isocratic, 10 min); Step 2, 0 to 2% Eluant B (linear, 5 min);Step 3, 2 to 9% Eluant B (linear, 6 min); Step 4, 9 to 13% Eluant B(linear, 6 min); Step 5,13 to 30% Eluant B (linear, 10 min); Step 6, 30 to0% Eluant B (linear, 5 min). The flow rate was set at 1 ml/min, and thecolumn temperature was set at 30°C. Nucleosides and nucleobases

used as optical markers for the test chromatograms were purchasedfrom Sigma Chemical Co. (St. Louis, MO). 3,/V4-Ethenodeoxycytidine andoxet-G were synthesized as described (14). W'.S-Ethenoguanine was a

generous gift of Dr. G. Doerjer (Institute of Toxicology, University ofMainz, Federal Republic of Germany) and was synthesized as described(16).

Analysis of Nucleotides. Nucleotides were resolved by anion-ex-change chromatography on a MicroPak AX-10 column (4 mm x 30 cm;

Varían,Saint Priest, France). The gradient consisted of Eluant A, water,and Eluant B, 11 mw orthophosphoric acid, pH 2.3 (analytical reagentgrade). The program was: Step 1, 20 to 100% Eluant B (linear, 8 min);Step 2, 100% Eluant B (15 min); Step 3, 100 to 20% Eluant B (linear, 5

min). Runs were made at ambient temperature and at a flow rate of 4ml/min.

Replication Fidelity Assays

Each reaction mixture (170 /il) contained 12.5 nmol untreated or CEO-treated poly(dG-dC) (nucleotide phosphorus); 400 pmol each of [8-3H]-dGTP (185 dpm/pmol), dCTP, dATP, or dTTP, S'-fa-^PldTTP (12,300dpm/pmol), or 5'-[a-Å“P]dATP (17,800 dpm/pmol) (free nucleoside 5'-

triphosphates were obtained from P-L Biochemicals; radiolabeled nu-

cleotides were from Amersham France, Versailles, France); 1 unit £coliMRE 600 DNA polymerase I (EC 2.7.7.7, Grade I; Boehringer Mannheim);7 nriMmagnesium chloride; and 50 mw Tris-HCI buffer (pH 7.8). After 1h incubation at 30°C, the solutions were spotted onto Whatman GF/C

glass microfiber filters (Whatman S.A., Ferneres, France), and DNA wasprecipitated in cold 5% (w/v) trichloroacetic acid. The filters were rinsed3 times in cold 5% trichloroacetic acid containing 2% (w/v) sodiumpyrophosphate, twice in 5% trichloroacetic acid alone, and once inethanol; the radioactivity present on the filters was determined by scintillation counting using Aquasol 2 (NEN Chemicals GmbH, Dreieich,Federal Republic of Germany).

Nearest-Neighbor Analysis

Template Mixture IX was used for nearest-neighbor analysis and for

nicking by an AP endonuclease.Replication of Control Template and Template IX, by E. coli ONA

Polymerase I. The incubation medium (1.30 to 1.35 ml) consisted of82.5 nmol control template or 95 nmol Template IX; 2 nmol each ofdGTP, dCTP, dTTP, or dATP, [KP]dATP, or [œP]dTTP; 20.4 units £coli

DNA polymerase I; 4.5 mw magnesium chloride; and 50 mw Tris-HCI

buffer (pH 7.8).Purification of the Replicated Templates. After 1 h incubation at

30°C, the polynucleotides were purified from excess nucleotides on

Elutip-d columns (Schleicher & Schuell GmbH, Dassel, Federal Republic

of Germany). Samples were charged on the columns and washed repeatedly with a low-salt solution [0.2 M sodium chloride:! ÕTIMEDTA:20rriM Tris-HCI buffer (pH 7.3 to 7.5)]. Templates were then eluted in ahigh-salt solution [1 M sodium chloride: 1 mM EDTA:20 mw Tris-HCI buffer(pH 7.3 to 7.5)], desalted on Sephadex G-50F (same conditions as under"Treatment of Templates"), and lyophilized.

Enzymatic Hydrolysis (10). Each of the 4 templates (control templateand Template IX labeled with S'-ta-^PJdAMP or with S'-fa-^PJdTMP

(13 to 25 ¿ig)was dissolved in 975 M<5 mivi Tris-HCI buffer (pH 8.6) with

2 mM calcium chloride. Micrococcal endonuclease (EC 3.1.31.1, fromStaphylococcus aureus, Grade VI; Sigma) was added at 125 ^g (11-8units) in 25 n\ Tris-HCI buffer, and incubation was carried out at 37°C

for 2 h. After the pH of each digest had been reduced to 6.85 with dilutedhydrochloric acid, 50 /ig (0.1 unit) calf spleen phosphodiesterase (EC3.1.16.1; Boehringer Mannheim) were added, and the mixture wasincubated at 37°C.At the end of 1 h, the same amount of enzyme was

added, and incubation was continued for another 2 h.Determination of the Specific Activities of 3'-dCMP and 3'-dGMP.

Enzymatic hydrolysates were brought to pH 2.3 by the addition of dilutedorthophosphoric acid and centrifugea for 5 min at 10,000 x g to removeany paniculate. The solutions were dried under a stream of nitrogen; theresidues were then redissolved in water for HPLC analysis, and theradioactivity was determined in the fractions.

Nicking of AP Sites

Micrococcus luteus AP endonuclease B (a generous gift from Drs. J.Pierre and J. Laval, Institut Gustave Roussy, Villejuif, France) was usedto nick phosphodiester bonds at AP sites (20). The reaction mixture (110/tl) contained 50 mM Tris-HCI buffer (pH 8.0), 2.5 mM magnesium chloride,6.25 nmol poly(dG-dC) (nucleotide phosphorus), and 1 or 2 units APendonuclease B. After 2 h at 37°C, the mixtures were heated at 50°C

CANCER RESEARCH VOL. 45 JUNE 1985

2441

on July 12, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

EFFECTS OF CEO-INDUCED LESIONS ON DNA REPLICATION

for 3 min. Controls were processed in the same way but without APendonuclease.

For replication fidelity assays, 400 pmol [in 10 n\ Tris-HCI buffer] eachof [3H]dGTP [185 dpm/pmol], dCTP, dATP, or dTTP, [^PJdTTP [5800dpm/pmol], or [Å“P]dATP [9800 dpm/pmol] and 2 units E. coli DNA

polymerase I [in 23 M! 50 rriM Tris-HCI (pH 7.8):40 mw magnesiumchloride] were added, and incubation was carried out for 1 h at 30°C.

Radioactivity incorporated into DNA was measured as described under"Replication Fidelity Assays."

RESULTS

HPLC Analysis of Nucleobase Adducts in CEO-treatedPoly(dG-dC). Treatment of poly(dG-dC) (double-stranded form)with increasing concentrations of CEO resulted in the dose-dependent formation of oxet-G as the main alkylation product inpoly(dG-dC) (Table 1, Column 2). Only trace amounts of 3,/V4-

ethenodeoxycytidine were formed after treatment with the highest concentrations of CEO (Templates VI, VII, Vili; Table 1,Column 3). A/'.S-Ethenoguanine could not be detected.

Misincorporation of dTMP and dAMP in CEO-treatedPoly(dG-dC) Templates. When poly(dG-dC) templates containing increasing amounts of oxet-G and eC were incubated in thepresence of E. coli DNA polymerase I and the 4 deoxyribonucle-oside 5'-triphosphates, the rate of replication decreased, and

the frequencies of dAMP and dTMP incorporation increased(Table 1, Columns 4, 5, and 6).

Nearest-Neighbor Analysis of Replicated Templates. In order to determine the sites of misincorporation of dAMP anddTMP, control poly(dG-dC) and Template IX were replicated withE. coli DNA polymerase I in the presence of [32P]dAT?, or [32P]-

dTTP and the 3 other nucleotides (Table 1). Following incubation,the templates were purified, hydrolyzed to nucleoside 3'-mono-

phosphates, and analyzed by HPLC and scintillation counting(Table 2). A major part of adenine was incorporated oppositecytosine, whereas thymine was distributed equally opposite cy-tosine and guanine; the 2 distributions varied by only 4 to 5%from one template to the other (Table 2, Columns 4 and 7). Thecontributions of cytosine and guanine lesions to the misincorporation events were evaluated by their mispair frequencies(Table 2, Columns 5 and 8) calculated as error rate (Table 1,Columns 5 and 6) multiplied by the fraction of CpY or GpYsequences (Table 2, Columns 4 and 7; Y stands for A or T). Thefrequencies of mispairings induced by CEO treatment werecalculated by subtracting the background frequencies (Table 2,Column 5) from the frequencies observed with Template IX(Table 2, Column 8): C:A, 13.2 x 1(T3; G:A, 3.3 x 10~3; C:T, 0.9x 1CT3;G:T, 1 x 1(T3. Thus, 80% of the CEO-induced dAMP

incorporation occurred opposite cytosine residues; but the proportions of new C:T and G:T mismatches were roughly equal.The contributions of cytosine and guanine lesions to the overalldAMP and dTMP misincorporations induced by CEO treatmentwere 77 and 23%, respectively; therefore, cytosine lesions wereabout 3 times more efficient than were guanine lesions in decreasing the fidelity of DNA synthesis.

Misincorporation of dAMP and dTMP following Nicking ofAP Sites. In order to assess the role of AP sites in the replica-tional errors observed with CEO-treated poly(dG-dC), Template

IX was nicked with M. luteus AP endonuclease B and replicatedby E. coli DNA polymerase I in the presence of the 4 deoxyri-bonucleoside 5'-triphosphates, including [3H]dGTP and either

pPJdATP, or pPJdTTP. After Template IX had been nicked byan AP endonuclease, the efficiency of DNA synthesis was enhanced. Taking the rate of replication of control poly(dG-dC) as

100% (see Table 1), the efficiency of replication of Template IX

Table1CEO-modifiedpoly(dG-dC) templates: HPLCanalysisand misincorporationassays

Poly(dG-dC)templates (Column 1) were obtained by reaction of poly(dG-dC)with various concentrations ofCEO. After purification, the polynucleotideswere hydrolyzed either enzymaticallyat neutral pH (in presenceofONaseI, snake venom phosphodiesterase.and acid phosphatase)or chemicallyat acidpH (in0.1 Nhydrochloricacid for 45 min at 75°C).The supematants containing nucleosides and/or nucteobases were analyzed byHPLC. DNA adducts were identified by their retention volume and quantitated by their absorbance at 254 nm(Columns 2 and 3). Each template was replicated in the presence of £coli DNA polymerase I, of equalconcentrations of the 4 nucleoside 5'-tripnosphates, and of Mg1'*. Incorporation of complementary andnoncornplementary nucleotides was measured by counting the radioactivity of |3H|dGMP (Column 4),[MP]dTMP (Column 5), and [MP]dAMP (Column 6) in acid-insoluble material; blank values, obtained in theabsenceof polymeraseor template, have been subtracted.

Template11poWdG-dC)1IIillIVVVIVIIViliIXeOxet-G:guanine"00.01

4±0.0020.047±0.0020.1

23±0.0060.185±0.0050.290

±0.0120.337±0.0090.381±0.0070.361±0.0120.360"eC:cytosinec0ND'NDNDNDND0.00250.0050.0050.005"dGMP

incorporation(pmol)128.9±12.6'48.2

±5.322.8±2.633.2±1.712.5

±1.611.6±1.63.23

±0.354.35±0.163.02

±0.603.53"10*x

errorrateddTMP:dGMP8.84.0148.77.31624293228"dAMP:dGMP121335575798150180200177"

* Roman numerals refer to poty(dG-dC)treated with various amounts of CEO.b Molar ratio, oxt- determined by HPLC, following acid hydrolysis.c Molar ratio, determined as nucleosidesby HPLC, following enzymatic hydrolysis.

Expressedas the noncomptementary:complementarynucleotidemolar ratio incorporatedon the templates.* Mean ±SD; n = 3.' ND, not detected.8 Mixture of equimolar amounts of Templates VI, VII, and VIII. These templates were pooled in order to

facilitate nearest-neighboranalyses(seeTable2). Composition, replicationrate, and error rates were calculatedas the means of the corresponding values observed with Templates VI, VII, and Vili.

CANCER RESEARCH VOL. 45 JUNE 1985

2442

on July 12, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

EFFECTS OF CEO-INDUCED LESIONS ON DNA REPLICATION

Tabte2Nearest neighbor analyses ol poly(dG-dC) templates and mispair frequencies

Control poly(dG-dC) and CEO-treated poly(dG-dC) (Template IX) were replicated ¡nthe presence of E coliDNA polymerase I and the 4 nudeoskte 5'-triphosphates, including [a-^PJdATP and Ia-MP]dTTP. Templateswere then hydrolyzed to nucleoside 3'-monophosphates. which were separated by HPLC and counted by

liquid scintillation. Fractions (Columns 4 and 7) of the nearest-neighbor sequences were calculated from thespecific activities of 3'-dCMP and 3'-dGMP (Columns 3 and 6). Each XpY sequence corresponds to a X:Y

mispair formed between template and daughter strand (X stands for C or G, Y for A or T). Data from themisincorporation assays (Table 1, Columns 5 and 6) were combined with data from nearest-neighbor analyses,and the mispair frequencies (Columns 5 and 8) were calculated as

X:Y mispair frequency = [XpY, nearest-neighbor sequence fraction] x [dYMP/dGMP, error rate].The error rates were: for poly(dG-dC), dAMP:dGMP = 1.2 x 10"3, dTMP:dGMP, 0.88 x 1CT3;for TemplateIX, dAMP:dGMP, 17.7 x liH, dTMP:dGMP, 2.8 x KT3.

5'-[«œP]Nu-cleotidedATPdTTPNearest-neighborsequenceCpA

GpACpT

GpTPoly(dG-dC)dpm/„g'4972

880589

511Fraction"0.85

0.150.54

0.4610*x

frequency01.0

0.20.50.4dpm/,91498382152150Template

IXfraction0.80

0.200.50

0.50103x

fre

quency14.2

3.51.4

1.4' dpm/Mg 3'-dCMP or 3'-dGMP.6 Fraction of radioactivity from [Å“P]dATP (or pPJoTTP) recovered in 3'-dCMP and 3'-dGMP, respectively.0 Mispair frequencies.

was 3%; it was 12% following incubation with 1 unit of APendonuclease and 20% when 2 units of enzyme were used.

The error rates of dTMP:dGMP and dAMP:dGMP were decreased by factors of 1.7 and 2.4, respectively, after incubationof Template IX with 2 units of endonuclease. There were differenteffects on dAMP incorporation and replication rate; the replication rate was doubled when 2 units of enzyme were used insteadof 1, but there was no further (or only a marginal) reduction inthe dAMPrdGMP ratio.

DISCUSSION

In HPLC analyses, following the reaction of CEO with poly(dG-dC) in a double-stranded form (Table 1), oxet-G amounted to

>98% of the total adducts identified. This finding is consistentwith those obtained by analysis of liver DNA from vinyl chloride-treated rodents (13,14,19). A^.S-Ethenoguanine and eC may be

formed from chloroacetaldehyde or CEO (1, 11, 13, 16); theformer adduct was not detected in our templates, and onlyminute amounts of <C were formed at the highest concentrationsof CEO. The presence of the hydrated intermediate of fC, anonmiscoding adduct (25), was not investigated. This differentialalkylation by CEO of guanine and cytosine moieties in double-

stranded DNA is consistent with its nucleophilic selectivity[Swain-Scott's constant, s = 0.83 (8)]; by comparison with

methyl methanesulfonate, which has a similar selectivity,CEO:cytosine adducts would be expected to comprise <1% ofthe CEO:guanine adducts (22).

It has been proposed that oxet-G is in equilibrium with a cyclichemiacetal form, O6,7-(1 '-hydroxyethano)guanine, after reaction

of the 7-(2-oxoethyl) group with the Opposition, and that thisform could, by analogy with O6-methylguanine (22), miscode for

thymine (Ref. 21; Chart 1). In order to test this hypothesis, weperformed misincorporation assays with CEO-treated poly(dG-dC) templates and E. coli DNA polymerase I. Because O6,7-(1 '-

hydroxyethano)guanine may represent only a small proportion ofthe total oxet-G residues, we used also highly modified templatescontaining up to 27% of oxet-G residues (Table 1). dTMP incorporation was enhanced only after treatment of poly(dG-dC) with

high concentrations of CEO, failing to support the above hypothesis.

Although an increase in dAMP incorporation was observedwith all the CEO-treated templates, nearest-neighbor analysis

(Table 2) demonstrated that lesions mainly at cytosine sites wereinvolved. Data from the misincorporation assays and from nearest-neighbor analyses show that at most 0.4 and 1.2% of theoxet-G moieties would mispair with thymine and adenine, respectively. Furthermore, about one-half of the dAMP and dTMP

misincorporations may be due to AP sites, as shown by theeffect of AP endonuclease treatment on the replication rate andon the error rates.

Our results indicate, therefore, that direct miscoding of oxet-

G may not contribute greatly to the induction of mutagenicity byCEO or vinyl chloride. Investigations are in progress to evaluatethe possible role of minor vinyl chloride:DNA adducts (1,13,17,18, 24).

ACKNOWLEDGMENTS

We wish to thank Dr. A. Croisy (Institut Curie, Orsay, France) for the preparationof CEO, Drs. J. Pierre and J. Uval (Institut Gustave Roussy, Viltejuif, France) forthe kind gift of AP endonuclease B, and J-C. Béréziatfor skillful technical assistance.We are indebted to M. Wrisez for typing the manuscript and to E. Heseltine foreditorial assistance.

REFERENCES

1. Barbin, A., and Bartsch, H. Mutagenic and promutagenic properties of DNAadducts formed by vinyl chloride adducts. In: H. Bartsch and B. Singer (eds.),The Role of Cyclic Nucleic Acid Adducts in Carcinogenesis and Mutagenesis.Lyon, France: International Agency for Research on Cancer, ¡npress, 1985.

2. Barbin, A., Bartsch, H., Lecomte, P., and Radman, M. Studies on the miscodingproperties of 1,N*-ethenoadenine and S.W-ethenocytosine. DNA reaction prod

ucts of vinyl chloride metabolites, during in vitro DNA synthesis. Nucleic AcidsRes., 9:375-387,1981.

3. Barbin, A., Brésil,H., Croisy, A., Jacqukjnon, P., Malaveilte, C., Montesano,R., and Bartsch, H. Liver microsome-mediated formation of alkylating agentsfrom vinyl bromide and vinyl chloride. Biochem. Biophys. Res. Commun., 67:596-603,1975.

4. Creech, J. L, and Johnson, M. N. Angiosarcoma of the liver in the manufactureof polyvinyl chloride. J. Occup. Med., 76:150-151,1974.

5. Green, T., and Hathway, D. E. Interactions of vinyl chloride with rat liver DNAin vivo. Chem.-Btol. Interact.. 22:211-224,1978.

6. Gwinner, L. M., Laib, R. J., Riser, J. G., and Bolt, H. M. Evidence of chtoreth-

CANCER RESEARCH VOL. 45 JUNE 1985

2443

on July 12, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

EFFECTS OF CEO-INDUCED LESIONS ON DNA REPLICATION

ylene oxide being the reactive metabolite of vinyl chloride towards DMA:comparative studies with 2,2 ' -dichlorodlethylether Cardnogenesis (Lond.) 4:

1483-1486.1983.7. HaH, J. A., Saffhill, R., Green, T., and Hathway, D. E. The induction of errors

during in vitro DNA synthesis following chloroacetaldehyde treatment ofpoly(dA-dT) and poly(dC-dG) templates. Carcinogenesis (Lond.), 2: 141-146,

1981.8. Hussein, S., and Oster man-Golkar. S. Comment on the mutagenic effective

ness of vinyl chloride metabolites. Chem.-Biol Interact., 12: 265-267,1976.

9. [ARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals toHumans, Vol. 19, pp. 377-438. Lyon, France: International Agency for Re

search on Cancer, 1979.10. Josse, J., and Swartz, M. Determination of frequencies of nearest-neighbor

base sequences in deoxynbonudeic acid. Methods Enzymoi. 6: 739-751,

1963.11. Kochetkov, N. K.. Shibaev, V. N., and Kost, A. A. New reaction of adenine and

cytosine derivatives, potentially useful for nucleic acids modification. Tetrahedron Lett., 22: 1993-1996,1971.

12. Laib, R. J. Specific covalent binding and toxicity of aliphatic halogenatedxenobiotics in A. H. Beckett and J. W. Gorrod (eds.). Reviews on DrugMetabolism and Drug Interactions, pp. 1-48. London: Freund Publishing

House, Ltd., 1982.13. Laib, R. J. The rote of cyclic base adducts in vinyl chloride induced carcino-

genesis: studies on nucleic acid aikyiation in vivo. In: H. Bartsch and B. Singer(eds.), The Rote of Cyclic Nucleic Acid Adducts in Carcinogenesis and Muta-genesis. Lyon, France: International Agency for Research on Cancer, in press,1985.

14. Laib, R. J., Gwmner. L. M.. and Bolt, H. M. DNA aikyiation by vinyl chloridemetabolites: etheno derivatives or 7-alkylation of guanine? Chem.-Biol. Interact., 37:219-231,1981.

15. Mattoni, C., Lefemine, G., Chieco, P., and Carretti. D. Vinyl chloride Carcinogenesis: current results and perspectives. Med. Lav., 65: 421-444,1974.

16. Oesch, F., and Doerjer, G. Detection of Af*,3-ethenoguanine in DNA aftertreatment with chloroacetaldehyde in vitro. Carcinogenesis (Lond.), 3: 663-

665,1982.17. Oesch, F., Adler, S., Rettelbach, R., and Doerjer, G. Repair of etheno DNA

adducts by N-glycosylases. In: H. Bartsch and B. Singer (eds.), The Rote of

Cyclic Nucleic Acid Adducts in Carcinogenesis and Mutagenesis. Lyon, France:International Agency for Research on Cancer, in press, 1985.

18. O'Neill, l., Barbin, A., Friesen, M., and Bartsch, H. Reaction kinetics and

cytosine adducts of chloroethylene oxide and chloroacetaldehyde. Direct observation of intermediates by FTNMR and GC-MS. in: H. Bartsch and B. Singer(eds.), The Rote of Cyclic Nucleic Acid Adducts in Carcinogenesis and Mutagenesis. Lyon, France: International Agency for Research on Cancer, in press,1985.

19. Osterman-Golkar, S., Hultmark, D., Segerback. D., Calteman, C. J., Gothe. R.,Ehrenberg, L., and Wachtmeister, C. A. Aikyiation of DNA and proteins in miceexposed to vinyl chloride. Biochem. Biophys. Res. Commun., 76: 259-266,1977.

20. Pierre, J., and Laval, J. Micrococcus lúteas endonucleases for apurinic/apy-nmidinic sites in deoxynbonucleic add. Biochemistry, 79: 5018-5029,1980.

21. Scheren E., Van der Laken, C. J., Gwinner, L. M., Laib, R. J., and Emmetot,P. Modification of deoxyguanosine by chloroethylene oxide. Carcinogenesis(Lond.), 2: 671-677,1981.

22. Singer, B., and Grunberger, D. (eds.) Molecular Biology of Mutagens andCardnogens. New York: Plenum Publishing Corp., 1983.

23. Singer, B., Abbott, L. G., and Spengler! S. J. Assessment of mutagenicefficiency of two cardnogen-modified nudeosides, 1,N*-ethenodeoxyadeno-sine and O*-methyldeoxythymidine, using polymerases of varying fidelity.

Cardnogenesis (Lond.), 5:1165-1171,1984.24. Singer, B., Hdbrcok, S. R., Fraenkel-Conrat, H., and Kusmierek. J. T. Neutral

reactions of haioacetaldehydes with polynudeotides: mechanisms, monomerand polymer products. In: H. Bartsch and B. Singer (eds.), The Rote of CyclicNucleic Acid Adducts in Cardnogenesis and Mutagenesis. Lyon, France:International Agency for Research on Cancer, in press, 1985.

25. Singer, B., Kusmierek, J. T., and Fraenkel-Conrat, H. In vitro discrimination ofreplicases acting on cardnogen-modified polynucleotide templates. Proc. Nati.Acad. Sd. USA, 80:969-972,1983.

26. Van Duuren, B. L., Katz, C., GokJschmidt, B. M., Frenkel, K., and Sivak, A.Caranogenicity of halo-ethers. II. Structure-activity relationship of analogs ofbis(chloromethyl) ether. J. NaM.Cancer Inst., 48:1431-1439,1972.

27. Zajdela. F., Croisy, A., Barbin, A., Malaveilte, C., Tomatis, L., and Bartsch, H.Carcmogenicity of chloroethylene oxide, an ultimate reactive metabolite of vinylchloride, and bis(chioromethyi) ether after subcutaneous administration and ininitiation-promotion experiments in mice. Cancer Res., 40: 352-356,1980.

CANCER RESEARCH VOL. 45 JUNE 1985

2444

on July 12, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1985;45:2440-2444. Cancer Res   Alain Barbin, Reinhold J. Laib and Helmut Bartsch  Major Vinyl Chloride-DNA AdductLack of Miscoding Properties of 7-(2-Oxoethyl)guanine, the

  Updated version

  http://cancerres.aacrjournals.org/content/45/6/2440

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/45/6/2440To request permission to re-use all or part of this article, use this link

on July 12, 2018. © 1985 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from