modulation of repair of 06-methylguanine in parenchymal and

[CANCER RESEARCH 45, 4768-4773, October 1985]

Modulation of Repair of 06-Methylguanine in Parenchymal and

Nonparenchymal Liver Cells of Rats Treated withDimethylnitrosamine1

Ghyslaine Planche-Martel, Alexei Likhachev,2 Christopher P. Wild,3 and Ruggero Montesano4

Unit of Mechanisms of Carcinogenesis, International Agency for Research on Cancer, 150 Cours Alben-Thomas, 69372 Lyon Cedex 08, France

ABSTRACT

Previous studies have shown that chronic treatment of ratswith dimethylnitrosamine (DMN) (2 mg/kg for 3 weeks) results inincreased repair of 06-methylguanine (06-mGua) in liver DNA.

The experiments reported here try to determine if this increasedrepair is confined to one or more cell populations of the liver.Liver cells [parenchymal (PC) and nonparenchymal (NPC)] wereseparated by elutriation centrifugation at various times after thelast administration of DMN.

The in vivo alkylation studies show that at any time after adose of [14C]DMN (2 mg/kg) the level of O6-mGua in PC cells of

DMN pretreated rats was much lower than in the same cellpopulation from control rats receiving only a single dose of DMN.In contrast, the pretreatment schedule resulted in no change inthe levels of this DNA adduct in NPC cells.

These results were confirmed by the determination of thelevels of O6-methyldeoxyguanosine by radioimmunoassay in

DNA from PC or NPC cells of rats similarly either pretreated for3 weeks with DMN or receiving a single dose of DMN (2 mg/kg).The in vitro measurements of 06-mGua DNA alkyltransferase

activity, using alkylated DNA as substrate, also show a higheractivity of this repair enzyme in PC cells. The DMN pretreatmentresulted in a 25-fold difference in O6-mGua DNA alkyltransferase

activity between the two cell populations of the liver.

INTRODUCTION

The study of the mechanisms underlying the specificity of thecarcinogenic action of the nitrosamines in a given animal species,organ, or cell type provides one of the most challenging areas ofresearch in experimental carcinogenesis. It both provides a valuable system for assessing the role of the various determinantsof the carcinogenic process, and is a source of biochemicalknowledge which can be valuable in the extrapolation of carcin-

ogenicity data across various species, including humans. Amongthe determinants that appear critical in the initiation of carcinogenesis by nitrosamines (1), it is DNA alkylation and subsequentDNA repair processes which seem the most relevant (2-5) and

in particular, repair of alkylation damage at the oxygen atoms ofpurine and/or pyrimidine bases (6, 7).

1Supported in part by Grant ENV-654-F from the Commission of the European

Communities.2 Present address: Laboratory of Experimental Tumours, Petrov Research In

stitute of Oncology, Pesochny 2, 188646 Leningrad, USSR.' Work carried out during the tenure of an IARC fellowship.4To whom requests for reprints should be addressed.5The abbreviations used are: DMN, dimethylnitrosamine; O'-mGua, O'-methyl-

guanine; 0*-mdGuo, O'-methyldeoxyguanosine; 7-mGua, 7-methylguanine; 3-

mAde, 3-methyladenine; PC, parenchymal; NPC, nonparenchymal; RIA, radioimmunoassay.

Received 2/25/85; revised 6/10/85; accepted 6/19/85.

In rats the inability of a single dose of DMN5 to induce tumors

in the liver (except when the nitrosamine is given after partialhepatectomy) is associated with the high capacity of this organto repair 06-mGua from its DNA (8, 9). In contrast, chronic

treatment with DMN results in the induction of liver tumors (10),and information from various large scale dose-response carcin-ogenicity studies carried out with DMN in rats are available (11-

13). Repeated administration of DMN to BD IV rats has beenshown to increase specifically the repair of O6-mGua in liver (14),

and this effect has now been observed under various experimental conditions (15). Previous studies (14, 16, 17) in ourlaboratories have shown that the increased repair of O6-mGua

from total liver DNA (a) was dependent, both upon the dose andthe length of DMN treatment; (b) persists for various weeks uponcessation of the treatment; and (c) this increased repair capacitywas limited to repair of relatively low levels of O6-mGua; inaddition it was shown that increased repair of O6-mGua does

not appear to be linked to increased DNA synthesis. All thesestudies were carried out using DNA from whole liver, while herewe report results obtained for the two liver cell populations, PCand NPC cells, considered individually under conditions whichwere found to be optimal for the induction of the increased repairof O6-mGua. Following DMN treatment the disappearance withtime of O6-mGua from the two cell populations In vivo was

measured either radiochemically or by RIA, and in parallel theO6-mGua DNA alkyltransferase activity of crude protein extracts

from PC or NPC cells was determined.

MATERIALS AND METHODS

Chemicals[14C]DMN (57.3 to 60.7 mCi/mmol) was synthesized in the Paterson

Laboratories, Manchester, United Kingdom, from [14C]dimethylamine

hydrochloride. Unlabeled DMN was obtained from Merck-Schuchardt,

Munich, Federal Republic of Germany. Collagenase type IV, obtainedfrom the Sigma Chemical Co., St. Louis, MO, and collagenase, obtainedfrom Boehringer Mannheim, France, were always tested for activitybefore use, due to marked batch variations. All the other compoundsused were of the purest grade available.

Experimental Design

The animals used in all experiments were male BD IV rats bred in theanimal house of the International Agency for Research on Cancer. Theanimals received food and water ad libitum, except that they were fastedovernight before perfusions.

Experiment 1. Rats weighing 110-150 g at the start of the experiment

received a daily dose on weekdays of 2 mg of unlabeled DMN/kg bystomach intubation for 3 weeks. The last dose on the 14th or 15thtreatment day of the experiment was 2 mg [14C]DMN/kg i.p. 24 h after

the last administration of unlabeled DMN to the above rats. The samedose of [14C]DMN (2 mg/kg i.p.) was administered to a group of control

rats not pretreated with unlabeled DMN. Liver perfusions and separations

CANCER RESEARCH VOL. 45 OCTOBER 1985

4768

Research. on February 2, 2018. © 1985 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

MODULATION OF REPAIR OF O8-mGua

of parenchymal and nonparenchymal cells by elutriation centrifugation(see below) were performed 2, 6, or 10 h after administration of theradioactive carcinogen. The cells were frozen in liquid N2 and stored at-80Â°C prior to DMA extraction. Results for each time point were obtained

from 2 or 3 rats and the variation observed was less than 20%.Experiment 2. Rats were pretreated with unlabeled DMN (2 mg/kg)

for 3 weeks as in Experiment 1, and 2, 6, or 24 h after the lastadministration of unlabeled DMN the livers were perfused and the PCand NPC cells were separated. Another group of rats received a singledose of unlabeled DMN by stomach intubation, and 2, 6, or 24 h laterthe two types of liver cells were isolated. Cells were frozen and storedat -80Â°C for subsequent analysis of 06-mdGuo content in DNA by RIA,

using a monoclonal antibody specific for this alkylated base. Theseantibodies were kindly provided by Drs. Boyle and Saffhill (PatersonLaboratories, Christie Hospital, and Holt Radium Institute, Manchester,United Kingdom) (18). For each time point the results were obtained from1-2 rats.

Experiment 3. Rats were pretreated with unlabeled DMN (2 mg/kg)for 3 weeks by stomach intubation as described above, and 2, 6, or 24h after the last administration of unlabeled DMN the livers were perfusedand the PC and NPC cells were separated. Control samples wereobtained from rats which received no treatment. Cells were frozen andstored as above. Cell extracts were subsequently prepared (see below)for determination of O6-mGua DNA alkyltransferase activity.

Liver Cell Separation

All rats, received an i.p. injection of 1250 IU heparin and were anesthetized with sodium phÃ©nobarbital prior to perfusion. Livers were perfused in situ by a collagenase technique adopted from Seglen (19), andthe resulting mixed liver cell suspension was separated into PC and NPCcells by elutriation centrifugation, using a modification of the methoddescribed by Lewis and Swenberg (20).

Cell debris and blood were removed from the NPC fraction by usingmetrizamide gradients and two purified populations were obtained. Cellyields per liver varied from 40 to 180 x 108 NPCs (which includeendothelial and Kupffer cells), and from 300 to 700 x 106 PCs with 96

to 98% and 80 to 85% viability, respectively. No difference was observedin cell yield from liver of control or pretreated rats. The whole separationprocedure took approximately 6 h, after which cells were frozen in liquidNZ and stored at -80Â°C until use.

In Experiments 1 and 2 DNA was isolated from PC and NPC cellsaccording to the procedure described by Umbenhauer and Pegg (21),with some modifications. At the end of the extraction from the PC pellet,DNA was redissolved in water and the solution was ultracentrifuged at30,000 rpm for 1 h to remove glycogen. In the case of NPC cells, onlyone extraction with chlorofornvisoamylalcohol (10:1; v/v) was used andthere was no incubation with Pronase. DNA was stored at -30Â°C untilanalysis. The yield of DNA was approximately 14 Mg/108 cells for PCand 8 Mg/106 cells for NPC.

In Experiment 1 the DNA was hydrolyzed to release free purine basesby heating at 70Â°C for 30 min in 0.1 N HCI. After addition of unlabeledmarkers of methylated purines (3-mAde, 7-mGua, and O6-mGua) the

sample was adjusted to pH 2.8-2.9, and then chromatographed on acolumn of Sephadex G-10 (90 x 0.6 cm) and eluted with 0.05 M ammo

nium formate containing 0.02% (w/v) sodium azide, pH 6.75. Theamounts of guanine and adenine in the sample were calculated from theUV absorption of the fractions containing these bases. The amounts ofmethylated bases present are expressed as mol/106 mol of the parent

base. Further details of this analytical procedure are given elsewhere(22).

The measurements by RIA in Experiment 2 were performed using thesystem and the monoclonal antibodies described by Wild ef al. (18). DNAextracted from PC and NPC cells was dissolved in buffer [10 mw TrisbufferÃ² mM MgCI2:1 /Â¿M2'-deoxycoformycin (National Cancer Institute,

Bethesda, MD)] for enzymic hydrolysis. Digestion was carried out with200 units DNase I/ml (Boehringer Mannheim) at 37Â°C for 2 h, pH 7,

followed by adjustment to pH 8 with 0.5 M Tris buffer. Digestion wascompleted by adding alkaline phosphatase, 170 units/ml (Boehringer)and snake venom phosphodiesterase, 0.075 units/ml (Boehringer Mannheim) and incubating overnight at 37Â°C. Enzymes were removed by

ethanol precipitation, and hydrolysates were air dried, reconstituted in0.8 ml 10 mM ammonium bicarbonate, and applied to an Aminex A7cation exchange separation system consisting of two columns (5 x 0.75cm) and (30 x 0.75 cm) in series. Separations were carried out at 50Â°C

using 10 mM ammonium bicarbonate, pH 8, as eluting buffer at a flowrate of 0.75 ml/min. The eluate was monitored at 254 nm and fractionscontaining individual parent deoxynucleosides were quantitated spectrc-photometrically. Fractions containing O6-mdGuo were pooled, air dried,

and reconstituted in 350 or 400 n\ of buffer (phosphate-buffered saline;

1 UM deoxycoformycin: 1% fetal calf serum: 3 mw sodium azide) for RIA.

Assay for O'-Methylguanine ONA Alkyltransferase

Cells were homogenized by vortexing in 1-2 ml of buffer (50 mw

Tris:0.1 mM EDTA:1.0 mM dithiothreitol, pH 7.5) and sonicated threetimes at 4Â°Cfor 30 s at 1-min intervals. The mixture was centrifuged at

12,000 x g for 10 min, the pellet was resuspended in 1 ml of the samebuffer, sonicated, and centrifuged as above. The combined supernatantswere centrifuged again at 12,000 x g for 30 min, and the proteinconcentration was determined by the Bio-Rad protein assay (23, 24). Allthe extracts were stored at -80Â°C until use.

The O6-mGua DNA alkyltransferase assay was carried out as de

scribed by Hall ef al. (25). The assay mixture, total volume 3 ml, alwayscontained 0.15 mg [meffty/3H]DNA (with 364 fmol of O6-mGua; 1300

dpm) but different amounts of protein (0.5 mg up to 20 mg). Afterincubation (37Â°Cfor 1 h) the reaction was stopped by the addition of 0.1

volume of saturated sodium acetate and 2 volumes of ice-cold ethanol.

The DNA was hydrolyzed in 0.1 M perchloric acid for 2 h, and the purinebases were separated, using a Sephadex G-10 column or a Partisil 10/25 SCX column when high-pressure liquid chromatography separation

was used. In each set of experiments a control was included in whichthe cell extract was incubated separately from the substrate DNA andmixed with it at the time of precipitation. For each sample the removalcapacity was calculated from the total dpm present as O6-mGua in the

control, and at the same time it was checked that no variations hadoccurred at the level of 7-mGua. Three to four rats were used individually

for each time point. The results are expressed as mean values of theseassays.

Carcinogenicity Experiments

Groups of 29 male BD IV rats (weighing 90-120 g at the start of the

experiment) were treated with a daily dose (2 mg/kg) of DMN by stomachintubation for 24 weeks (5 days/week) and were observed up to 120weeks of age. Groups of 29 control rats received saline and werefollowed for the same period of time.

RESULTS

In Vivo Alkylation in Parenchymal and NonparenchymalLiver Cells. Table 1 shows the levels of 3-mAde, O6-mGua, and7-mGua in DNA of PC and NPC liver cells at various times aftera dose of [14C]DMN (2 mg/kg) to control rats, or to rats that had

previously received unlabeled DMN at a dose of 2 mg/kg onweekdays for a period of 3 weeks. In both cell types the levelsof 3-mAde at all three time points were very similar in control

and pretreated rats. Also, no significant differences were observed in the 7-mGua levels, although a slightly higher level was

detected at 2 h in both cell types of the pretreated rats as


4769




Table 1

Alkylated purines in DNA of parenchymal and nonparenchymalcells at various 0.10times following administration of [ifC]DMN

Alkylated purines (mol/108mol of the base) were measured in DNA of PC andNPCcells at various times following i.p. administrationof [14C]DMN(2 mg/kg; 57.3 mto 60 mCi/mmol) to BD IV rats treated with DMN (2 mg/kg on weekdays by ~ Q08stomach intubation) for 3 weeks. Control rats received [14C]DMNonly. â„¢

3-Methyladenine Â°'-Methylguan- 7.Metny|guanine ^

CelltypePCNPCTime(h)26102610Pretreated241410271610Control181811211713Pretreated1696573739Control374830366555Pretreated697659643606426443Control569775610441653555

compared to controls; this effect was not evident at 6 or 10 h.In addition, there was a small effect in NPC cells where a lowerlevel of 7-mGua was observed compared to PC cells in both

control and pretreated rats at 2, 6, and 10 h. However, in thecase of O6-mGua, marked differences were observed between

the two cell populations, in that the level of this adduct in theDNA of PC cells from pretreated rats 2 h after the dose of [14C]DMN was less than one-half (16 mol/106 mol of guanine) thelevel observed in PC cells from control rats (37 mol/106 mol of

guanine); even greater differences were detected at 6 and 10 h.In contrast, in the DNA of NPC cells no such difference in O6-

mGua level between pretreated and control rats was observedand in fact, at 2 h a higher level was observed in pretreatedversus control rats (Table 1).

The differences in the relative amounts and the persistence ofO6-mGua in the two types of liver cells from pretreated or controlrats are also evident in Chart 1, showing the O6-mGua:7-mGua

ratios at 2, 6, and 10 h. These ratios, based on the radioactivitypresent in different fractions from the same column eluate, areprobably a more accurate representation of the observed differences, since the individual measurement of alkylated basesdepends on the determination of the actual amount of the parentbase present in DNA, and the quantities obtained from NPC cellswere in some instances very small. As shown in Chart 1 the O6-

mGua:7-mGua ratios in PC cellular DNA from pretreated ratswere one-third the corresponding ratios in control rats at 2, 6,and 10 h; in DNA from NPC cells no significant differences areevident between the two groups of rats. In this Chart the O6-

mGua:7-mGua ratios observed previously (16) in total liver DNA

in groups of rats similarly treated are shown for comparison. Itis evident that the specific increased repair of O8-mGua in

pretreated rats is confined to the PC cell population of the liver.The specificity of this effect for the O6-mGua lesion is further

indicated by the fact that the other two DNA alkylation productsmeasured (3-mAde and 7-mGua) are not affected by the DMN

pretreatment (see Table 1).In Experiment 2 the levels of O6-mdGuo were determined by

RIA in DNA from PC or NPC liver cells of rats pretreated for 3weeks with DMN or receiving a single dose of DMN (2 mg/kg).The RIA was used as it allowed a greater sensitivity thanradioactive determinations, and this proved especially importantfor measurements with NPC cells where, as mentioned above,much less DNA was available for analysis. In agreement with thestudies using [14C]DMN (Table 1), the results from the RIA

0.06

0.04

0.02

10 24Time(hr)after[I4C]DMNadministration

Chart 1. O*-Methylguanine:7-methylguanineratios in DNA of PC cells (â€¢,O),NPCcells(A, A), and total liverDNA(â€¢.Ãœ).at various times following administrationof [14C]DMN(2 mg/kg) to BD IV rats treated with unlabeled DMN at a dose of2 mg/kg on weekdays for 3 weeks (O, A, D). Control rats (â€¢,A, â€¢)received[14C]DMN(2 mg/kg) only.

Table 2Concentrationsof O'-methyldeoxyguanosine(mol O'-methyldeoxyguanosine/KP

mol deoxyguanosine)in DNAof parenchymal and nonparenchymalcells,measuredby radioimmunoassayat various times following last dose of DMNBD IV rats were treated with DMN(2 mg/kg on weekdaysby stomach intubation)

for 3 weeks. Control rats received only one dose of DMN (2 mg/kg by stomachintubation).

Time(h)2

624Parenchymal

cellsPretreated107

2Control6820 4Nonparenchymal

cellsPretreated37

3618Control364035

analysis (Table 2) also showed that the levels of O8-mdGuo in

PC cells from rats receiving pretreatment were much lower, at 2and 6 h posttreatment, than those seen in the same cell population from rats receiving only a single dose of DMN. Even 24 hafter the final dose of DMN, a time point not examined inExperiment 1, the level of O6-mdGuo in the pretreated groupwas one-half that found in the controls. The 06-mdGuo mea

surement for NPC cells, from both pretreated and control rats,again showed no differences between the two groups at 2 and6 h, emphasizing that the discrepancy between pretreated versuscontrol rats is limited to the PC cell population only. There wassome evidence of lower 06-mdGuo levels in NPC cells from

pretreated rats at 24 h when compared to controls (see alsoTable 1), but this small difference requires further investigationto determine whether it is a real difference between the twogroups.

Table 3 shows the incorporation at 14Cfrom [14C]DMN into

the normal adenine base via the C-1 pool. No difference in

incorporation was observed in PC cells from pretreated andcontrol rats, whereas in NPC liver cells a 20-fold increase com

pared to PC cells was detected in control rats, and a still higherincorporation in pretreated rats. This indirectly suggests that amuch higher cell replication occurs in NPC cells after treatment


4770




with DMN (2 mg/kg), and is in accord with the increased cellreplication which was observed in NPC cells of F344 rats measured by [3H]thymidine and bromodeoxyuridine labeling followingchronic administration of 1,2-dimethylhydrazine (26).

O8-Methylguanine ONA Alkyltransferase Activity in Paren

chyma! and Nonparenchymal Cells. The results obtained in PCand NPC liver cells of the O6-mGua DMA alkyltransferase activi

ties are directly complementary to the in vivo findings. Chart 2shows that in control rats a 50% repair of 06-mGua from theDMA (containing 364 fmol O6-mGua/150 /*g DNA) occurs with 4x 106 PC cells, whereas 13.5 x 106 NPC cells were required for

similar removal. In the rats pretreated with 2 mg DMN/kg for 3weeks in PC liver cells assayed 24 h after the last DMN injection,50% repair occurs with 1 x 106 cells; in NPC cells no increased

activity was observed following DMN treatment, and if anythinga lower activity was detected versus the corresponding NPCcontrol cells. The results are expressed per viable cell (and notper mg protein), as this is a more correct representation of theresults, since the amount of protein per cell greatly differs between the two cell populations analyzed (24, 27). The differencein the O6-mGua DNA alkyltransferase between PC and NPC cells

in control rats is very similar to that observed in another strainof rats (24). The DMN pretreatment resulted in a 25-fold difference in O6-mGua DNA alkyltransferase activity between the two

cell populations of the liver (Table 4).Carcinogenicity Studies. Chronic administration of daily

doses of 2 mg of DMN/kg for a total of 24 weeks resulted in 25tumor-bearing animals, of which 24 developed tumors of liver;these were 19 hepatocellular carcinomas and 7 hemangioen-dothelial tumors. One rat had a glioma and a mesenchymal tumorof the kidney, but no liver tumors. In a group of 29 control rats,no liver tumors were observed.

DISCUSSION

The aim of the present and previous studies (14,16,17) wasto examine the modulation of DNA repair processes dealing with

Table 3

Metabolic incorporation into adertine measured 6 h after i.p. administration of["CJDMN (2 mg/kg; 57.3 to 60.7 mCi/mmol) to BDIV rats treated with DMN (2

mg/kg on weekdays by stomach intubation) for 3 weeks.Control rats received [14C]DMN only

Cell type Pretreated Control

PCNPC106 4080117 2312

various DNA alkylation adducts during chronic administration ofnitrosamines.

The initial level (2 h) of 7-mGua was slightly higher in PC

compared to NPC cells (Table 1). In order to alkylate DNA, it isproposed (28) that DMN has to be metabolized through anoxidative demethylation to the unstable nitrosohydroxy-

methylmethylamine which generates a methyl carbonium Â¡on.Itis thus probable that in vivo the metabolic activation of DMNoccurs in liver PC cells, and the methylating species crosses thecellular membrane and reacts with DNA of nonhepatic cells ofthe liver, which are limited in or lack completely this metaboliccompetence. This is consistent with the finding of Umbenhauerand Pegg (29), showing that freshly isolated hepatocytes areable to metabolize DMN, and that the intermediatenitrosohydroxymethylmethylamine, with an estimated half-life of

10 s. (30), is able to alkylate extracellular DNA. Thus althoughsome differences in initial alkylation are observed (Table 1) (31),the difference in susceptibility of the various cell populations ofthe liver should be considered in terms of two events occurringsubsequent to carcinogen interaction with cellular DNA repairand DNA synthesis. In pretreated rats, both NPC and PC cellsshow the highest 7-mGua levels 2 h after exposure as compared

to 6 h in control rats. The data suggest that chronic exposure toDMN causes some induction of metabolizing enzymes as wellas modulating repair, as discussed below.

Treatment of rats with 2 mg DMN/kg for a period of 3 weeksresults in an increased repair of 06-mGua, but not of 7-mGua or

3-mAde, in DNA obtained from the liver as a whole (16), and this

effect persists for several weeks after cessation of the DMNtreatment (17). This specific increase of 06-mGua repair in liver

DNA is confined to the PC cells. In fact Chart 1 shows that thedifference between pretreated and control rats in the O6-mGua:7-

mGua ratios in total DNA or in the DNA from PC cells is of thesame order of magnitude (approximately 4- to 5-fold). The O6-

mGua DNA alkyltransferase activity also confirms (see Chart 2)that in control rats PC cells have approximately four times higheractivity than in NPC cells, and this difference increases to 25-

fold following pretreatment with DMN. Calculation of the numberof molecules of O6-mGua DNA alkyltransferase per cell (55,000

for PC and 7,800 in NPC liver of control rats and 110,200 and4,500 in the DMN-treated rats) are very close to those reported

in the literature (5). Similarly, in BD IX and F344 rats (20, 24, 31)chronic treatment with 1,2-dimethylhydrazine resulted in an increased repair of 06-mGua only in the PC cells, and the magni

tude of the increased repair was of the order observed in our

Chart 2. O6-Methylguamne DNA alkyltransferase activity (expressed as percentage of O'-

methylguanine removed from methylated DNAcontaining 364 fmol in 150 (/g DNA) in PC cells(â€¢,O) and NPC cells (A, A) of BD IV rats 24 hafter cessation of continuous exposure to DMN(2 mg/kg for 3 weeks). Control rats (â€¢,A)received no DMN. Regression lines were drawnfrom the values obtained from 3.5 x 105 to 3 x107 cells.

= 80

60

20

Parenchymalcells Nonparenchymalcells 364

S182 .g

S0

10s

Viablecell number


4771




Table 4O'-Methylguanine DNA alkyltransferase activity in parenchyma! and

nonparenchymal cells of control rats and rats treated with DMNO'-Methylguanine DNA alkyltransferase activity was measured in PC and NPC

extracts of BD IV rats at 24 h after cessation of treatment with DMN (2 mg/kg onweekdays for 3 weeks). Control rats received no DMN. Each value represents themean Â±SD of data from 3 rats. Methylated DNA (150 ng) contained 364 fmol(1300 dpm) of [3H]O<>-methylguanine.

Activity in PC Activity in NPC

fmol/h/mg fmol/h/10* fmol/h/mg fmol/h/106protein cells protein cells

ControlTreated71

Â±23148 Â±4291

Â±42183Â±64179

Â±27113Â±512

Â±27.5 Â±0.3

experiments.In the experiments quantitated by RIA we are measuring total

06-mdGuo, including both alkylation from pretreatment sched

ules as well as from the final dose of DMN. One might expect tofind therefore an accumulation of this lesion in NPC cells uponchronic exposure, as the O6-mGua repair capacity is apparently

not increased sufficiently to remove all of the lesion in 24 h. Theresults reported here however show no difference of O6-mdGuo

levels in pretreated NPC cells compared to controls 2 h after thefinal dose of DMN (Table 2). This suggests that any 06-mdGuo

present due to accumulation is small in comparison to thequantity of this adduci induced by such a dose of this carcinogen.This observation is in accord with that of Bedell ef al. (31), whereaccumulation of O6-mdGuo/mg DNA in NPC cells was seen up

to 8 days during daily exposure to 1,2-dimethylhydrazine, but

that the concentration of the adduci in DNA declined markedlyover the next 3 weeks. It has been reported that the rate of denovo DNA synthesis in NPC cells is higher than in PC cells incontrol rats, and also is increased far more in the former celltype compared to the latter in response to daily exposure to 1,2-

dimethylhydrazine (26). We measured metabolic incorporation of[14C]DMN into adenine via the C-1 pool and found 20-fold and

40-fold more incorporation in control and pretreated NPC cells

compared to PC control and pretreated cells, respectively (Table3). Thus a large increase in DNA synthesis specifically in NPCcells would lead to a dilution of the concentration of O6-mdGuo

in this cell population, resulting in an apparent reduction ofaccumulation. Under these circumstances it might be arguedthat it is difficult to eliminate the possibility that part of the lackof 06-mdGuo accumulation in NPC cells is a reflection of a smallinduction of O6-mGua DNA alkyltransferase activity masked bythe "dilution" effect discussed above. The situation would be

made clearer from a knowledge of the level of 7-mGua or 3-

mAde, which were not determined in this instance. However thedata from the alkyltransferase studies (Chart 2) show no evidence of increased repair in NPC cells and again strongly suggestthat such an increase is limited to the PC cell population.

In rats treated with 1,2-dimethylhydrazine and in mice after

DMN, liver hemangioendothelial tumors are more frequently observed, hepatocellular tumors however are also reported (31,32). The prevalence of liver tumors originating from NPC cellshas been associated with the low capacity of these target cellsto repair 06-mGua and the concomitant high level of DNA syn

thesis induced by the carcinogen treatment (24, 31). The doseof DMN and the duration of treatment in the carcinogenicityexperiments reported here however, in which mainly hepatocellular carcinomas were induced, were too high to allow detection

of any possible difference in carcinogenic susceptibility betweenPC and NPC cells. It should be stressed in all cases that theseconsiderations be taken in the context of the stochastic processof carcinogenesis, and that the contrasting carcinogenic response of the different liver cell populations is not so clear-cutand is greatly affected by the dose of carcinogen. For examplea recently reported (13) large scale study in Colworth rats showsthat the induction by DMN of hepatocellular carcinomas is approximately linear in the dose range below 0.1 mg/kg/day, andan approximately cubic relationship is observed within the higherrange of doses. This "hockey stick" carcinogenic dose response

could be possibly related to the findings reported here andpreviously (14,16,17) of an increased repair of 06-mGua in DNA

of hepatocytes after chronic treatment with relatively low dosesof DMN, and in the saturation of this DNA repair process whenhigh doses are used (33, 34). It is also evident from the experiments in Colworth rats that this repair process, although efficientand assuming that it is directly related to the frequency of cancerinduction by nitrosamines, cannot reduce to zero the probabilityof developing tumors. In Escherichia coli the induction of O6-

mGua DNA alkyltransferase is of a much higher level, going from20 to 10,000 molecules/cell, and resulting in a more drasticreduction of mutation frequency after a challenging dose of W-methyl-AT-nitro-iV-nitrosoguanidine (35, 36).

Swenberg ef al. (37) observed in rats an accumulation of O4-ethyldeoxythymidine but not of 06-ethyldeoxyguanosine in DNA

of hepatocytes following treatment with diethylnitrosamine,which induces a high incidence of hepatocellular carcinomas.This result points to O4-ethyldeoxythymidine as a more critical

DNA lesion in this instance, and suggests that these two eth-ylation DNA adducts may be repaired by different enzymes or, ifindeed the same DNA alkyltransferase is involved, then it has amuch higher efficiency for removal of O6-ethyldeoxyguanosinethan of O4-ethyldeoxythymidine. In E. coli the purified O6-mGuaDNA alkyltransferase repairs both O6-mGua and O4-methyldeox-

ythymidine (38), and in mammalian cells there is also strongevidence that the same DNA alkyltransferase acts on bothlesions although with a different affinity (39). This is an importantpoint that requires clarification and it should also be noted thatother 0-alkylated pyrimidines may be involved in initiation, but

studies are as yet limited by lack of sensitive assay methods.Specificity at a cell population level in relation to initial DNAdamage and subsequent DNA repair processes is certainly relevant to nitrosamine carcinogenesis; however an aspect thathas received so far little attention is the possible specificity ofdistribution of these DNA adducts within chromatin structure ofthe cellular genome. Recently Nehls ef a/. (40) have shown theoccurrence of highly sensitive sites for O8-ethylguanine adducts

in brain DNA following in vivo treatment of rats with ethylnitro-

sourea. This observation is of particular interest since it wasobserved (41, 42) that in mammary tumors induced in rats by N-methyl-A/-nitrosourea the activation of H-ras oncogene was theresult of a single point mutation (Gâ€”A transition), which isprecisely what one would expect from a defective repair of O6-mGua. Both these observations imply the occurrence of "hotspots" of alkylation within the DNA which are of relevance to the

carcinogenesis process.

ACKNOWLEDGMENTSWe would like to thank O. Deblock for her technical help, Dr. R. Becker for his

criticisms, Dr. J. R. P. Cabrai for his consultation on the pathology, and P. Collard-Bianchi for her secretarial help.


4772




REFERENCES

1. Bartsch, H., and Montesano, R. Relevance of nitrosamines to human cancer.Carcinogenesis (Lond.), 5: 1381-1393,1984.

2. Goth, R., and Rajewsky, M. F. Persistence of O6-ethylguanine in rat-brainDNA: correlation with nervous system-specific carcinogenesis by ethylnitro-sourea. Proc. Nati. Acad. Sci. USA, 77: 639-643,1974.

3. Lawley, P. D. Carcinogenesis by alkylating agents. Chemical Carcinogens.ACS Monogr., 773: 83-244, 1976.

4. Montesano, R. Alkylation of DNA and tissue specificities in nitrosaminescarcinogenesis. J. Cell. Biochem., 77: 259-273, 1981.

5. Pegg, A. E. Repair of O8-methylguanine in DNA by mammalian tissues. In: H.

Greim, R. Jung, M. Kramer, H. Marquardt, and F. Oesch (eds.). BiochemicalBasis of Chemical Carcinogenesis, pp. 265-274. New York: Raven Press.

1984.6. Pegg, A. E. Methylation of the O'-position of guanine in DNA is the most likely

initiating event in carcinogenesis by methylating agents. Cancer Invest., 2:223-231,1984.

7. Singer, B. Alkylation of the 0" of guanine is only one of many chemical events

which may initiate carcinogenesis. Cancer Invest., in press, 1985.8. Craddock, V. M. Uver carcinomas induced in rats by single administration of

dimethylnitrosamine after partial hepatectomy. J. Nati. Cancer Inst., 47: 889-

907,1971.9. Rabes. H. M., and Szymkowiak, R. Cell kinetics of hepatocytes during the

preneoplastic period of diethylnitrosamine-induced liver carcinogenesis. Cancer Res., 39: 1298-1304, 1979.

10. Magee, P. N., Montesano, R., and Preussmann, R. W-Nitroso compounds andrelated carcinogens. Chemical carcinogens. ACS Monogr., 773: 491-625,

1976.11. Terracini, B., Magee, P. N., and Barnes, J. M. Hepatic pathology in rats on low

dietary levels of dimethylnitrosamine. Br. J. Cancer, 27: 559-565,1967.12. Arai, M., Aoki, Y., Nakanishi, K., Miyata, Y., Mon, T., and Ito, N. Long-term

experiment of maximal non-carcinogenic dose of dimethylnitrosamine for carcinogenesis in rats. Gann, 70: 549-558,1979.

13. Peto, R., Gray, R., Brantom, P., and Grasso, P. Nitrosamine carcinogenesis in5120 rodents: chronic administration of sixteen different concentrations ofNDEA, NDMA, NPYR and NPIP in the water of 4440 inbred rats with parallelstudies on NDEA alone of the effect of age of starting (3, 6 or 20 weeks) andof species (rats, mice or hamsters). In: I. K. O'Neill, R. C. Von Borstel, C. T.

Miller, J. Long, and H. Bartsch (eds.), N-Nitroso Compounds, Occurrence,Biological Effects and Relevance to Human Cancer, IARC Scientific PublicationNo. 57, pp. 627-665. Lyon, France: International Agency for Research on

Cancer, 1985.14. Montesano, R., BrÃ©sil,H., and Margison, G. P. Increased exicison of O*-

methylguanine from rat liver DNA after chronic administration of dimethylnitrosamine. Cancer Res., 39:1798-1802, 1979.

15. Margison, G. P., Cooper, D. P., Smith, R. A., Montesano, R., BrÃ©sil,H.,Planche-Martel, G., and O'Connor, P. J. Enhanced repair of O'-alkylguanine

in mammalian tissues. Folia Bioi. (Prague), 132-142, Special Publication, 1984.16. Montesano, R., BrÃ©sil,H., Planche-Martel, G., Margison, G. P., and Pegg, A.

E. Effect of chronic treatment of rats with dimethylnitrosamine on the removalof O'-methylguanine from DNA. Cancer Res., 40: 452-458,1980.

17. Montesano, R., BrÃ©sil,H., Planche-Martel, G., Margison, G. P., and Pegg, A.E. Stability and capacity of dimethylnitrosamine-induced O'-methylguaninerepair system in rat liver. Cancer Res., 43: 5808-5814,1983.

18. Wild, C. P., Smart, G., Saffhill, R., and Boyle, J. M. Radioimmunoassay of 0s-

methyldeoxyguanosine in DNA of cells alkylated in vitro and in vivo. Carcinogenesis (Lond.), 4:1605-1609,1983.

19. Segten, P. O. Preparation of isolated rat liver cells. In: D. M. Prescott (ed.),Methods in Cell Biology, Vol. 13, pp. 29-83. New York: Academic Press, Inc.,

1976.20. Lewis, J. G., and Swenberg, J. A. Differential repair of Oe-methylguanine in

DNA of rat hepatocytes and nonparenchymal cells. Nature (Lond.), 288. IBS-

IS?, 1980.21. Umbenhauer, D. R., and Pegg, A. E. Alkylation of intracellular and extracellular

DNA by dimethylnitrosamine following activation by isolated rat hepatocytes.

Cancer Res., 41: 3471-3474,1981.22. Margison, G. P., Margison, J. M., and Montesano, R. Accumulation of Os-

methylguanine in non-target tissue DNA during chronic administration of dimethylnitrosamine. Biochem. J., 765: 463-468,1977.

23. Pegg, A. E., and Hui, G. Formation and subsequent removal of O'-methyl-

guanine from deoxyribonucleic acid in rat liver and kidney after small doses ofDMN. Biochem. J.. 773: 739-748, 1978.

24. Swenberg, J. A., Bedell, M. A., Billings, K. C., Umbenhauer, D. R., and Pegg,A. E. Cell specific differences in O6-alkylguanine DNA repair activity duringcontinuous exposure to carcinogen. Proc. Nati. Acad. Sci. USA, 79: 5499-

5502,1982.25. Hall, J., BrÃ©sil,H., and Montesano, R. Oe-Alkylguanine DNA transferase activity

in monkey, human and rat liver. Carcinogenesis (Lond.), 6: 209-211,1985.26. Lewis, J. G., and Swenberg, J. A. Effect of 1,2-dimethylhydrazine and dieth-

ylnitrosamine on cell replication and unscheduled DNA synthesis in target andnontarget cell populations in rat liver following chronic administration. CancerRes., 42: 89-92,1982.

27. Decker, K., Hofmann, F., Kreush, J., Maier, K. P., Munder, P. G., and Wagte,S. R. Nucleotide contents and nitrogen metabolism of rat Kupffer cells andperitoneal macrophages. In: E. Wisse and D. L. Knook (eds.), Kupffer Cellsand Other Liver Sinusoidal Cells, pp. 315-324. Amsterdam: Elsevier/North

Holland BiomÃ©dical,1977.28. Pegg, A. E. Metabolism of N-nitrosodimethylamine. In: R. Montesano, H.

Bartsch, and L. Tomatis (eds.), Molecular and Cellular Aspects of CarcinogenScreening Tests, IARC Scientific Publication No. 27, pp. 3-22. Lyon, France:International Agency for Research on Cancer, 1980.

29. Umbenhauer, D. R., and Pegg, A. E. Metabolism of DMN and subsequentremoval of 0*-methylguanine from DNA by isolated rat hepatocytes. Chem.-Biol. Interact.. 33: 229-238, 1981.

30. Mochizuki, M., Anjo, T., and Okada, M. Isolation and characterization of W-alkyl-N-(hydroxymethyl)nitrosamines from N-alkyl-W-(hydroperoxymethyl)ni-trosamines by deoxygenation. Tetrahedron Lett., 27: 3693-3696,1980.

31. Bedell, M. A., Lewis, J. G., Billings, K. C., and Swenberg, J. A. Cell specificityin hepatocarcinogenesis: preferential accumulation of O*-methylguanine intarget cell DNA during continuous exposure of rats to 1,2-dimethylhydrazine.Cancer Res., 42: 3079-3083,1982.

32. Lindamood, C., Bedell, M. A., Billings, K. C., and Swenberg, J. A. Alkylationand efe novo synthesis of liver cell DNA from C3H mice during continuousdimethylnitrosamine exposure. Cancer Res., 42: 4153-4157,1982.

33. Kleihues, P., and Margison, G. P. Exhaustion and recovery of repair excisionof 0Â«-methylguanine from rat liver DNA. Nature (Lond.), 259: 153-155,1976.

34. Pegg, A. E. Dimethylnitrosamine inhibits enzymatic removal of 06-methylgua-nine from DNA. Nature (Lond.), 274: 182-184,1978.

35. Caims, J., Robins, P., Sedgwick, B., and Talmud, P. The induciate repair ofalkylated DNA. Prog. Nucleic Acid Res. Mol. Biol., 26: 237-244, 1981.

36. Lindahl, T. DNA repair enzymes. Annu. Rev. Biochem., 57: 61-87,1982.37. Swenberg, J. A., Dyroff, M. C., Bedell, M. A., Popp, J. A., Huh, N., Kirstein,

U., and Rajewsky, M. F. O4-Ethyldeoxythymidine, but not 0*-ethykteoxygu-

anosine, accumulates in hepatocyte DNA of rats exposed continuously todiethylnitrosamine. Proc. Nati. Acad. Sci. USA, 87:1692-1695,1984.

38. McCarthy, T. V., Karran, P., and Lindahl, T. Inducible repair of 0-alkylatedDNA pyrimidines in Escherichia coli. EMBO J., 3: 545-550,1984.

39. Becker, R. A., and Montesano, R. Repair of O4-methyldeoxythymidine residuesin DNA by mammalian liver extracts. Carcinogenesis (Lond.), 6: 313-317,1985.

40. Nehls, P., Rajewsky, M. F., Spiess, E., and Werner, D. Highly sensitive sitesfor guanine-O*-ethylation in rat brain DNA exposed to N-ethyl-W-nitrosourea invivo. EMBO J.. 3: 327-332,1984.

41. Sukumar, S., Notario, V., Martin-Zanca, D., and Barbacid, M. Induction ofmammary carcinomas in rats by nitroso-methyl-urea involves the malignantactivation of the H-ras 1 locus by single point mutations. Nature (Lond.), 306:658-661,1983.

42. Notario, V., Sukumar, S., Santos, E., and Barbacid, M. A common mechanismfor the malignant activation of ras oncogenes in human neoplasia and inchemically induced animal tumors. In: G. F. Vande Woude, et al. (eds.), CancerCells. Vol. 2, pp. 425-432. Cold Spring Harbor, NY: Cold Spring HarborLaboratory, 1984.


4773



1985;45:4768-4773. Cancer Res Ghyslaine Planche-Martel, Alexei Likhachev, Christopher P. Wild, et al. DimethylnitrosamineNonparenchymal Liver Cells of Rats Treated with

-Methylguanine in Parenchymal and6OModulation of Repair of

Updated version

http://cancerres.aacrjournals.org/content/45/10/4768

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

[email protected] 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/45/10/4768To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



modulation of repair of 06-methylguanine in parenchymal and

Documents