- Email: [email protected]
Correlation of DNA adduct levels with tumor incidence: carcinogenic potency of DNA adducts
Mutation Research 424 Ž1999. 237–247
Correlation of DNA adduct levels with tumor incidence: carcinogenic potency of DNA adducts Michael Otteneder, Werner K. Lutz
)
Department of Toxicology, UniÕersity of Wuerzburg, Versbacher Strasse 9, D-97078 Wuerzburg, Germany Accepted 11 October 1998
Abstract The quantitative relationship between DNA adducts and tumor incidence is evaluated in this review. All available data on DNA adduct levels determined after repeated administration of a carcinogen to rats or mice have been compiled. The list comprised 27 chemicals, of all major structural classes of carcinogens. For the correlation with tumor incidence, the DNA adduct levels measured at the given dose were normalized to the dose which resulted in a 50% tumor incidence under the conditions of a 2-year bioassay ŽTD50 dose.. In rat liver, the calculated adduct concentration ‘responsible’ for a 50% hepatocellular tumor incidence spanned from 53 to 2083 adducts per 10 8 nucleotides, for aflatoxin B1, tamoxifen, IQ, MeIQx, 2,4-diaminotoluene, and dimethylnitrosamine Žin this order.. In mouse liver, the respective figures were 812 to 5543 adducts per 10 8 nucleotides, for ethylene oxide, dimethylnitrosamine, 4-aminobiphenyl, and 2-acetylaminofluorene. The observed span Ž40-fold in rats, 7-fold in mice. reflects differences between the various DNA adducts to lead to critical mutations. If additional carcinogens fit in with this astonishingly narrow range, the measurement of DNA adduct levels in target tissue has the potential to be not only an exposure marker but an individual cancer risk marker. For toremifen and styrene, low levels of DNA adducts were detected in rat liver at the end of a negative long-term bioassay. This shows that the limit of detection of DNA adducts can be well below the limit of detection of an increased tumor incidence. For a cancer risk assessment at low levels of DNA damage, treatment-related adducts must be discussed in relation to the background DNA damage and its inter- and intraindividual variability. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Cancer risk; Carcinogen; DNA adduct; Dose–response relationship; Neoplasm
1. Introduction DNA adducts are constantly being formed, even in the absence of exposure to genotoxic carcinogens X
Abbreviations: dG, 2 -Deoxyguanosine; MeIQx, 2-Amino3,8-dimethyl-imidazow4,5-fxquinoxaline; NNK, 4-Ž N-Methyl-NX nitrosamino.-1-Ž3-pyridyl.-1-butanone; 8-oxodG, 8-Oxo-2 -deoxyguanosine ) Corresponding author. Tel.: q49-931-201 5402; Fax: q49931-201 3446; E-mail: [email protected]
ŽRef. w1x, and references therein.. Humans get cancer even if they avoid exposure to the established cancer risk factors. Similarly, laboratory animals kept under well-controlled ‘carcinogen-free’ conditions show a spontaneous tumor incidence w2x. The question therefore is whether, and to what extent, background DNA adducts could be responsible for the spontaneously arising tumors. The efficiency of the ‘background DNA damage’ to induce ‘spontaneous tumors’ is unknown. The relationship between DNA adduct levels and tumor
0027-5107r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 2 7 - 5 1 0 7 Ž 9 9 . 0 0 0 2 2 - 6
238
Table 1 DNA adduct levels determined after repeated administration of carcinogens Carcinogen
Speciesr Sex
(A) Methylating and ethylating carcinogens N-Methyl-N-nitrosourea ŽMNU. Rm
Route
Dose Žmg kgy1 dy1 .
Duration
Site
Adduct concentration Žper 10 8 Nt.
Adduct types
Method
Reference
f34
wat
33.3 a
2w
640 1300 4100 46 1016 724 407 44 180 550 840 2500 175000 60000 250 7 10 38r3880 –r24120 120r35970 94r6900 1940r10000 190r9300 875r10200 2660r24800 8r28r78 3r13r39
dG-O 6-Me
Immunoassay
w10x
dG-O 6 -Me
Immunoassay
w11x
dG-O 6 -Me
HPLCrfluorescence and immunoassay
w12x
dG-O 6 -Mer dG-N 7 -Mer dT-O 4-Me dG-O 6 -Me dG-O 6 -Me
HPLCrfluorescence and immunoassay
w13x
Immunoassay Repair assayr ada protein HPLCrfluorescence
w14x w15x
N-Methyl-N -nitro-Nnitrosoguanidine ŽMNNG.
Rm
wis
wat
3.3 a
12 w
4-Ž N-Methyl-N-nitrosamino.-1Ž3-pyridyl.-1-butanone ŽNNK.
Rm
f34
ipj
12 d
Rm
f34
ipj
0.3 1 10 30 100
liv col for for stg duo jej lun
12 d
liv
Rm Rf
f34 wis
ipj wat
14 d 28 d
lun liv
Rm
f34
wat
16 d
liv
Mm
c3h
wat
16 d
liv
Mm
cbl
wat
16 d
liv
Procarbazine
Rf
sda
eat
10 d
liv lyd
Dacarbazine
Rf
sda
ipj
40 0.223 a 0.372 a 0.7 1.9 3.9 1.9 a 4.8a 1.8 4.8 9.5 5 10 20 20
10 d
1,2-Dimethylhydrazine ŽSDMH. Diethylnitrosamine ŽDEN.
Rm Rm
f34 f34
wat wat
2.2 0.03 a 0.08 0.3 0.8 3.3 8.3
16 d 70 d
liv lyd liv liv
X
Dimethylnitrosamine ŽDMN.
84 39 11300 3 3 42 112 224 224
dG-O 6-Mer dG-N 7 -Me
w16x
dG-O 6 -Mer dG-N 7 -Me dG-O 6 -Mer dG-N 7 -Me
HPLCr fluorescence w16x
dG-O 6 -Me
Repair assayr ada protein
dG-O 6 -Me
w18x Repair assayr ada protein HPLCr fluorescence w19x w20x Immunoassay
dG-O 6 -Mer dG-N 7 -Me
HPLCr fluorescence w16x
w17x
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
Strain
Rf
f34
eat
Toremifen ŽTOR.
Rf
f34
eat
Ethylene oxide ŽETO.
Mm
b6c
inh
Aflatoxin B1 ŽAFB1.
Rm
f34
wat
Styrene
Rm R mqf
wis cdr
2-Amino-3-methylimidazow4,5-fxquinoline ŽIQ.
300 610 7 13 7 1016 718
Six addrRAL
Postlab.
w4x
Six addrRAL
Postlab.
w4x
dG-N 7 -HOEt
HPLCr fluorescence
w21x
liv
0.1 3.2 85
Total radioact.
3
w22x
23 d 2y
liv liv
20 90
Total radioact. dG-O 6 -SO
3
H postlab.
w23x unpubl.
28 d
liv
dG-C8-ABP
postlab.
w24x
28 d
liv
dG-C8-ABP
postlab.
w24x
42 d
liv
dG-C8-MeIQx
liv
RAL, 7 spots
accel. mass spectrometry postlab.
w25x
12 w
w26x
15 d
liv sto liv sto liv
1170 4880 7470 9100 220 1220 2520 4400 0.0019 0.5 1 10 100 1000 200 50 750 100 500
RAL, add.stand.
postlab.
w27x
RAL, add.stand.
postlab.
w28x
18 m
liv
18 m
liv
28 d Ž5 drw.
lun liv
2.2=10y6 73=10y6 y3 2.1=10
8w
gav inh
1=10y3 1030
bal
wat
Mm
bal
wat
Rm
sda
eat
Rm
f34
eat
Rm
f34
gav
2.7 10.7 21.4 42.9 2a 7.8 15.7 31.4 9.5=10y6 5=10y3 0.007 0.065 0.672 6.3 5
(C) Aromatic amines and nitroarenes 4-Aminobiphenyl Mf Ž4-ABP.
2-Amino-3,8-dimethylimidazow4,5-fxquinoxaline ŽMeIQx.
a
20.8 41.7 20.8 a 41.7 62.5 b 67
a
50 Rm
f34
gav
100
10 d
HrHPLC
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
(B) Adduct formation Õia epoxide Tamoxifen ŽTAM.
239
240
Table 1 Žcontinued. Carcinogen
Speciesr Sex
Route
Dose Žmg kgy1 dy1 .
Duration
Site
Adduct concentration Žper 10 8 Nt.
Adduct types
Method
Reference
f34
gav
100
10 d
liv
50
RAL, add.stand.
postlab.
w28x
f34 f34 wf
ipj eat eat
postlab. immunoassay immunoassay
w29x w30x w31x
eat
dG-C8-AF
postlab.
w32x
2-Acetylaminophenanthrene Ž2-AAP.
R mqf
sda
ipj
4w
w33x
Rm
cdr
eat
50 a
5w
3
w34x
2-Nitrofluorene ŽNF.
Rm
wis
eat
33
10 d
dG-C8-AP dG-N 2 -AP 2 spots dG-C8-MAB dG-N 2 -MAB dA-N 6 -MAB dG-C8-AF
postlab.
N-Methyl-4-aminoazobenzene ŽMAB.
liv kid ubl liv spl kid liv kid for
330 7200 7400 7900 153r353 563r816 909r2094 1228r3425 1081r3750 1616r4675 1934r6831 2690r8966 200 122 131 900 150 50 11 6 54
3 spots, RAL dG-C8-AF dG-C8-AF
bal
10 d 28 d 31 d 46 d 28 d
liv liv liv
Mf
5 16.7 a 16.7 a 33.3 0.7 a 2.1 4.3 6.4 8.6 10.7 14.2 21.4 0.7 Ž1= wk.
(D) Polynuclear aromatic hydrocarbons Benzow axpyrene ŽBP. Mm
cbl
ipj
1
10 d
Benzow k xfluoranthene ŽBF.
Mm
cbl
ipj
1
10 d
Cyclopentaw cd xpyrene ŽCPP.
Mm
cbl
ipj
1
10 d
Fluoranthene ŽFA.
Rm
sda
eat
20.8 Ž13d. 41.7 Ž24d. mean: 27 a
37 d Ž13q24.
liv lun liv lun liv lun liv kid lun
3 2 3 8 12 36 49 48
Rf Rm
wis wis
gav gav
0.05 0.05
42 d 42 d
liv liv
36.5 0.6
(C) Aromatic amines and nitroarenes 2-Amino-1-methyl-6-phenylimidazo R m w4,5-fxpyrimidine ŽPhIP. 2,4-Diaminotoluene ŽDAT. R n.d. 2-Acetylaminofluorene Ž2-AAF. Rm Rm
(E) Miscellaneous Cyproterone acetate ŽCPA.
livrubl
HrHPLC
postlab.
w35x
RAL
postlab.
w36x
RAL
postlab.
w36x
RAL
postlab.
w36x
dN-anti-FADE, RAL
postlab.
w37x
4 spots, RAL 4 spots, RAL
postlab. postlab.
w38x
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
Strain
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
incidence could be investigated in experiments involving chronic exposure to genotoxic chemical carcinogens, such as the 2-year rodent bioassay for carcinogenicity. If a strong quantitative association can be shown for exogenous carcinogens, the observed levels of background adducts could be correlated with the control tumor incidence. The data available so far are compiled, analyzed, and discussed in this paper. 2. DNA adduct levels after repeated administration of carcinogens
241
analysis of a correlation between adduct levels and tumor incidence because the relationship will strongly depend on the time of analysis after the administration. The best data could be obtained if adduct levels were determined at the end of a standard bioassay for carcinogenicity. Unfortunately, such data are usually not available, with the exception of tamoxifen and toremifen Ž18-month treatment. w4x and the unpublished data for styrene mentioned below Ž2-year treatment.. As a compromise, DNA adduct levels reached after repeated administration for at least 10 days were included in the analysis.
2.1. DNA-adduct leÕels as a function of time
2.2. Literature search for DNA adduct data after repeated administration of carcinogens
The concentration of adducts measured in a DNA sample is the result of a balance between rates of formation, removal by repair and cell death, and ‘dilution’ by DNA replication Žthe newly synthesized strand does not contain adducts.. After a long period of treatment, a steady state is expected to be reached, except for repair-resistant adducts in non-dividing cells. For adducts that are rapidly repaired, steadystate levels are dominated by repair; for adducts that are refractory to repair, the cell turnover will be the rate-limiting factor. Most published DNA-adduct data have been determined after a single dose of a carcinogen w3x. These data are of limited value for a quantitative
Two Searches in Medline yielded 344 and 378 hits, of which 30 references contained valuable data on 34 experiments. Adduct levels derived from dermal administration were not included. A minimum duration of treatment for 10 days was required for the reference to be evaluated. Table 1 summarizes the results. Data were available for carcinogens from a variety of classes, 7 methylating agents and 1 ethylating agent Žpart A., 5 agents which alkylate via an epoxide Žpart B., 7 aromatic amines and 1 nitroarene Žpart C., 4 polynuclear aromatic hydrocarbon Žpart D., and 1 steroid analogue Žpart E.. Most data were available for the liver. Other tissues such as lung, kidney and
Notes to Table 1: Species: M, mouse; R, rat. Sex: f, female; m, male; n.d., no data. Strain: bal, BALBrc; b6c, B6C3F1 ; cbl, C57BL; c3h, C3H; cdr, CD Charles River; f34, F344; sda, Sprague–Dawley; wf, WF; wis, Wistar. Route: eat, diet; gav, gavage; inh, by inhalation; ipj, intraperitoneal injection; wat, in drinking water. Dose: a The concentration in the drinking water or in the feed Žexpressed in ppm. was converted to mg kgy1 dy1 by division by 12 Žrat. or 7 Žmouse.. b The concentration in the air Žexpressed in ppm. was converted to mg kgy1 dy1 on the basis of a ventilation rate of 0.03 l mousey1 miny1 Ž0.25 lrrat. and 100% absorption. The underlined values were used for the calculation of adduct levels in Table 2. Duration: d, day; w, week; m, month; y, year. Site: col, colon; duo, duodenum; for, forestomach; jej, jejunum; kid, kidney; liv, liver; lun, lung; lyd, lymph node; sto, stomach; stg, stomach, glandular; ubl, urinary bladder; Adduct concentration: Nt, nucleotide; conversion of units expressed per parent base was based on a molecular percentage of 22% G and C, 28% A and T. The underlined values were used for the calculation of adduct levels in Table 2. Adduct type: add, adduct; dA-N 6 , N 6-alkylated deoxyadenosine; dG-O 6 , O 6-alkylated deoxyguanosine; dG-N 2 , N 2-alkylated deoxyguanosine, dG-C8, C8-alkylated deoxyguanosine; dG-N 7, N 7-alkylated deoxyguanosine; dT-O 4 , O 4-alkylated deoxythymidine; Et, Ethyl; FADE, 2,3-dihydroxy-1,10b-epoxy-1,2,3-trihydro-fluoranthene; Me, Methyl; RAL, relative adduct labeling; SO, styrene oxide; total, total measured DNA-adduct level. Method: accel. mass spectrometry, accelerator mass spectrometry; 3 H, administration of 3 H-radiolabeled carcinogen; 3 H rHPLC, nucleotide analysis by HPLC after administration of 3 H-radiolabeled carcinogen; postlab., 32 P-postlabeling; repair assayrada protein, competitive repair assay using E. coli O 6-methylguanine–DNA alkyltransferase.
242
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
forestomach were investigated occasionally, depending on the organotropism for tumor induction by the carcinogen in question. Reported DNA adduct levels spanned one hundred million-fold from as low as about 2 adducts per 10 11 rat liver nucleotides, measured 42 days after a daily oral dose of 9.5 ng MeIQx per kg body weight Žadducts analyzed by accelerator mass spectrometry. to as many as 2 adducts per 1000 nucleotides in rat liver DNA, measured after 12 days of 100 mg NNK per kg body weight Žadduct analysis by HPLCrfluorescence.. DNA-reactive carcinogens usually result in multiple adducts. The number can span from a few Že.g., aromatic amines. to more than a dozen Že.g., small methylating agents such as the methyl diazonium ion CH 3 –Nq' N derived from dimethylnitrosamine.. Different adducts can have different consequences. For instance, while the methylation of the nitrogen atom a7 of guanine facilitates depurination, the methylation of guanine at the O 6 results in mispairing with T, thereby producing a G–C to A–T transition mutation after 2 rounds of DNA replication. Methylating agents which produce different ratios of N 7- vs. O 6- methylation therefore are expected to have different effects, possibly also with respect to tumor induction. The number of detected adducts depends on the method used. If the analysis is based on the determination of radioactivity after the administration of appropriately radiolabeled test chemicals, all adducts formed might be detected Že.g., aflatoxin B1.. On the other hand, if the analysis is based on photometric or mass spectrometric detection, it will be adductspecific. With the 32 P-postlabeling method, a quantification on the basis of RAL values Žrelative adduct labeling. could result in an underestimation of the actual adduct levels. These limitations are expected to affect the attempted quantitative correlation of adduct levels with tumor incidence data. 3. DNA adduct levels associated with a 50% liver tumor incidence 3.1. Normalization of adduct leÕels to the tumorigenic dose 50% (TD50 ) If all adducts had the same carcinogenic potency, the same adduct levels would be required for a given
tumor incidence. In order to test this hypothesis, the adduct levels compiled in Table 1 were normalized to a standard dose with respect to the induced tumor incidence. The TD50 value was chosen for this purpose. This is the dose which halves the probability to stay tumor-free within two years, or approximately, it is the dose which results in a 50% tumor incidence within 2 years. Criteria to calculate TD50 values have been set by Lois Gold and coworkers and all values have been published in the carcinogenic potency database which is available in print w5x and on the Internet Žhttp:rrpotency.berkeley.edurcpdb.. A number of criteria had to be met for a carcinogen and a study listed in Table 1 to be included for further evaluation. First, the dose used for the DNAadduct experiment should not be more than 10-fold higher than the TD50 value. When different dose levels were used for the DNA adduct determinations, the dose which was closest to the TD50 was used. Second, only those situations were included in the evaluation where the major or the majority of the adducts were analyzed. Third, in view of the scarcity of adduct data for organs other than the liver, only this organ was considered. With respect to histopathological criteria, the TD50 value referring to hepatocellular tumors was used, if available. Published TD50 values were available for most of the chemicals which met these criteria. For tamoxifen citrate, an unpublished value was communicated by Dr. Lois Gold. 3.2. Adducts Õs. tumor incidence The adduct concentrations calculated for the dose levels associated with a 50% liver tumor incidence are shown in Table 2. Proportionality was used for the adjustment, i.e., adduct concentrations given in column a8 of Table 1 were divided by the dose shown in column a5 and multiplied with the TD50 value shown in column a4 of Table 2. The result is shown in column a6 of Table 2. In the rat Žpart A., values span from 53 adducts per 10 8 nucleotides for the aflatoxin B1 to 2082 per 10 8 nucleotides for dimethylnitrosamine. The data could be interpreted to mean that the aflatoxin–DNA adducts are about 40 times more potent for the induction of hepatocellular carcinoma than the adducts measured with dimethylnitrosamine. In view
Table 2 DNA adduct concentrations in the liver calculated for the dose which results in a 50% hepatic tumor incidence ŽTD50 . in rat ŽA. and mouse ŽB.
(A) Rat Dimethylnitrosamine ŽDMN. Tamoxifen ŽTAM. Aflatoxin B1 ŽAFB1. 2-Amino-3,8-dimethyl-imidazow4,5-fxquinoxaline ŽMeIQx. 2-Amino-3-methylimidazow4.5fxquinoline ŽIQ. 2,4-Diaminotoluene ŽDAT. B. Mouse Dimethylnitrosamine ŽDMN. Ethylene oxide ŽETO. 4-Aminobiphenyl Ž4-ABP. 2-Acetylaminofluorene Ž2-AAF.
Sex
Strain
TD50 Žmg kgy1 dy1 .
Histopathology
Adduct concentration at TD50 Žper 10 8 Nt.
Comments
m f m m
f34 n.d.Ž ™f34. f34 f34
0.372 5.54 0.0013 1.26
bht Ž?. hpc hpc
2082 80 53 188
Major adducts RAL Total adduct level RAL
m
f34
2.08
hpc
83
RAL
fqm Ž ™ n.d..
f34
14
MXA q MXB
924
RAL
m f Ž ™ m. f m f
cbl b6c cif Ž ™ bal. cif Ž ™ bal. bal
0.179 75.8 3.05 7.5 44.1
mal hpa mhp mhp hpc
944 812 1322 1173 5543
Major adducts Single major adduct Single major adduct Single major adduct Single major adduct
Histopathology: bht, benign hepatocellular tumor; hpa: hepatocellular adenoma; hpc: hepatocellular carcinoma; hpt: hepatoma; mal: malignant tumor; mhp: malignant hepatoma; MXA: more than one tumor type, combined by NCIrNTP; MXB: more than one tumor type, combined by the authors of the CPDB at Berkeley. SexrStrain: The information given in the brackets refer to the DNA-adduct data.
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
Carcinogen
243
244
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
of the enormous span of DNA adduct levels compiled in Table 1, this is an astonishingly small span. Within a class of structurally similar carcinogens, for instance aromatic amines, the range is even smaller. It ranges from 83 to 924 per 10 8 nucleotides, i.e., spans only one order of magnitude. This means that adduct levels measured in the liver might indeed have some potential to predict tumor incidence. In the mouse Žpart B of Table 2., only 4 data sets fulfilled the criteria for analysis. The values were higher than in the rat and ranged from 812 to 5543 per 10 8 nucleotides, i.e., spanned only seven-fold. Whether the mouse liver can indeed tolerate larger levels of DNA adducts than the rat without tumor response is an interesting question. Spontaneous liver tumor formation normally is more pronounced in mice, as compared with rats. Although this is strainand sex-dependent, it could indicate that nongenotoxic aspects of carcinogenesis in the mouse liver will have to be given particular attention. 4. Discussion 4.1. Carcinogenic potencies of DNA adducts Based on the data available so far, the range of carcinogenic potency of structurally different DNA adducts is less than 2 orders of magnitude. Within a class of chemically related carcinogens, it is only 1 order of magnitude. This result refers to the liver and the induction of hepatocellular tumors. More data will have to be generated before the same analysis can be performed for other organs. Furthermore, the data are expected to be cell type-specific. In the liver, about 70% of the cells are hepatocytes. DNA isolated from a homogenate of this organ therefore is predominantly from the target cell type. In organs such as the lung or the kidney, an average level of DNA adducts determined in an organ homogenate is unlikely to represent the situation prevailing in a specific target cell population. For such situations, adduct data will have to be collected from enriched cell populations. 4.2. Measurable adduct leÕels in the absence of increased tumor incidence (example: styrene) The correlation established between adduct levels and tumor incidence also allows to discuss the ques-
tion whether an adduct concentration might be measurable without giving rise to a detectable increase in tumor incidence. With 50 animals per group as used in a standard bioassay for carcinogenicity and assuming a 5% background tumor incidence in the controls, a 5% increase in a dose group would not be statistically significantly different from the controls. A 5% increase is one-tenth of the incidence used to normalize the adduct levels. Therefore, an adduct level of up to 200 per 10 8 nucleotides in the rat liver and up to 500 per 10 8 nucleotides in mouse liver could go unobserved with respect to an increased tumor incidence. This upper limit is compatible with unpublished data obtained from liver DNA of rats treated for 2 years by inhalation with styrene. The main intermediate metabolite of styrene is styrene-7,8-oxide. This epoxide has weakly electrophilic activity and has a plasma half life of about 20 min in rats. Low levels of DNA adducts are therefore expected in all tissues. In some qualitative contrast, there was no evidence of carcinogenic activity in a recent 2-year bioassay for carcinogenicity of styrene by inhalation exposure of CD-rats to 50, 200, 500 and 1000 ppm, for 6 hrday and 5 daysrweek. Liver tissue was available from the rats at the end of the 2-year exposure. The 32 P-postlabeling technique was optimized for the detection and quantification of the 2 regioisomeric 2X-deoxyguanosyl-O 6-adducts from styrene-7,8-oxide w6x, and the liver DNA was analyzed. In the 1000 ppm group, total adduct levels were about 100 adducts per 10 8 nucleotides. This level is below the value mentioned above, expected to result in a statistically significant increase in tumor incidence. It shows that the limit of detection of DNA adduct levels can be much lower than the limit of detection for a significant increase in tumor incidence. In this example, the 32 P-postlabeling technique was used. Under optimal conditions, limits of detection of one adduct per 10 9 nucleotides can be achieved. This is equivalent to 20 adducts per diploid cellular genome containing 6 pg DNA. Using accelerator mass spectrometry, the adduct levels determined after MeIQx exposure corresponded to not even 1 adduct per cell. In view of the background DNA damage present at all time, it goes without saying that some adduct levels associated with an external exposure can be mea-
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
sured but cannot constitute a cancer risk worth mentioning. 4.3. Background DNA damage and spontaneous tumor incidence The question shall now be addressed whether the background DNA adduct level determined in the rat liver is compatible with the observed spontaneous liver tumor incidence. In most rat strains, the latter is in the low percent range. Assuming a spontaneous tumor incidence of 5%, an upper limit adduct level of 200 per 10 8 nucleotides could be associated with this tumor incidence. For I-compound adducts and for adducts derived from alkylating agents, aldehydes, and aromatic amines, this number would probably not be exceeded w1x, indicating that the background DNA adducts do not differ from the adducts formed by the exogenous chemical carcinogens with respect to carcinogenic potency. For DNA oxidations, the situation is critically dependent on the choice of reference for the levels of 8-oxo-deoxyguanosine and on the question of the levels of other oxidation products. On the basis of the recently published values for rat liver Ž0.23 8-oxodG per 10 5 dG, i.e., 50 8-oxodG per 10 8 nucleotides w7x., the level of 200 damaged nucleotides per 10 8 nucleotides would not be exceeded. However, taking into account that 8-oxodG is only one of a number of oxidative DNA lesions, the data could be also interpreted to mean that the oxidative background DNA damage is less dangerous than other types of DNA damage. Evolutionary pressure against oxygen-related damage might have resulted in the formation of specific, replication-coupled repair systems for oxidative DNA lesions. 4.4. Correlation between DNA adduct leÕels and cancer risk The use of DNA adduct data from chronic carcinogen exposure for cancer risk assessment has been discussed before by Poirier and Beland w8x. These authors focused on the shape of the dose-response curves for DNA adduct levels and tumor incidence in various organs, and compared the respective data for animals with humans. As an exam-
245
ple, the relationship between DNA adduct levels and bladder tumor incidence induced in male mice by 4-aminobiphenyl was compared with the situation in human smokers. The authors concluded that, per unit dose, humans may be more sensitive than mice to tumor induction by aromatic amine exposure andror that factors in addition to carcinogen exposure are contributing to the tumorigenic end point. A number of critical assumptions had to be made for this conclusion, in particular with respect to the adduct levels in the human bladder. It is hoped that biomonitoring data will accumulate in the future so that the uncertainties can be reduced. The question of a correlation of a DNA adduct level with tumor incidence on an individual basis has been addressed experimentally. Using the mouse skin tumor model and chronic exposure to 7,12-dimethylbenzw axanthracene ŽDMBA., the individual latency time to the appearance of the first papilloma was correlated with the DMBA–DNA adduct levels in the treated normal skin areas. Surprisingly, the adduct levels were lowest in those mice which showed the shortest latency period w9x. The explanation came from the analysis of the rate of cell division in the epidermis. The mice with the fastest tumor induction had the highest rates of cell division. Since the DNA adduct concentration is reduced by a factor of 2 with each round of DNA replication, cell division must be considered a confounder for the interpretation of DNA adduct levels. Our analysis showed that adducts derived from a variety of genotoxic chemical carcinogens differed by 1–2 orders of magnitude with respect to their carcinogenic potency in rodent liver. This range is small if compared with the range of possible exposures. Therefore, for a specific organ, the determination of a DNA adduct level could become predictive for a cancer risk. What is needed are more data on adduct levels in target organs of animals and humans. This would allow to interpret DNA adduct data not only as exposure markers but individual risk markers.
Acknowledgements We thank Dr. Lois Gold for making unpublished TD50 values available. Liver samples of styrene-
246
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
treated rats were a gift of the Styrene Information and Research Center ŽSIRC.. The styrene adduct analysis was supported by the Styrene Steering Committee of CEFIC.
References w1x R.C. Gupta, W.K. Lutz, Background DNA damage from endogenous and unavoidable exogenous exposure to genotoxic carcinogens, Mutat. Res. Ž1999., in press. w2x J.K. Haseman, J. Huff, G.A. Boorman, Use of historical control data in carcinogenicity studies in rodents, Toxicol. Pathol. 12 Ž1984. 126–135. w3x W.K. Lutz, In vivo covalent binding of organic chemicals to DNA as a quantitative indicator in the process of chemical carcinogenesis, Mutat. Res. 65 Ž1979. 289–356. w4x D. Li, Y. Dragan, V.C. Jordan, M. Wang, H.C. Pitot, Effects of chronic administration of tamoxifen and toremifene on DNA adducts in rat liver, kidney, and uterus, Cancer Res. 57 Ž1997. 1438–1441. w5x L.S. Gold, E. Zeiger, Handbook of Carcinogenic Potency and Genotoxicity Database, CRC Press, Boca Raton, FL 1997. w6x M. Otteneder, E. Eder, W.K. Lutz, 32 P-postlabeling analysis of DNA adducts of styrene-7,8-oxide at the O 6-position of guanine, Submitted to Chem. Res. Toxicol. 12 Ž1999. 93–99. w7x T. Hofer, L. Moller, Reduction of oxidation during the ¨ X preparation of DNA and analysis of 8-hydroxy-2 -deoxyguanosine, Chem. Res. Toxicol. 11 Ž1998. 882–887. w8x M.C. Poirier, F.A. Beland, DNA adduct measurements and tumor incidence during chronic carcinogen exposure in animal models: implications for DNA adduct-based human cancer risk assessment, Chem. Res. Toxicol. 5 Ž1992. 749–755. w9x W. Fischer, W.K. Lutz, Correlation of individual papilloma X latency time with DNA adducts, 8-hydroxy-2 -deoxyguanosine, and the rate of DNA synthesis in the epidermis of mice treated with 7,12-dimethylbenzwaxanthracene, Proc. Natl. Acad. Sci. U.S.A. 92 Ž1995. 5900–5904. w10x H. Ohgaki, B.I. Ludeke, I. Meier, P. Kleihues, W.K. Lutz, C. Schlatter, DNA methylation in the digestive tract of F344 rats during chronic exposure to N-methyl-N-nitrosourea, J. Cancer Res. Clin. Oncol. 117 Ž1991. 13–18. w11x N.H. Zaidi, C.S. Potten, G.P. Margison, D.P. Cooper, P.J. O’Connor, Tissue and cell specific methylation, repair and synthesis of DNA in the upper gastrointestinal tract of Wistar X rats treated with N-methyl-N -nitro-N-nitrosoguanidine via the drinking water, Carcinogenesis 14 Ž1993. 1991–2001. w12x S.A. Belinsky, C.M. White, T.R. Devereux, J.A. Swenberg, M.W. Anderson, Cell selective alkylation of DNA in rat lung following low dose exposure to the tobacco specific carcinogen 4-Ž N-methyl-N-nitrosamino.-1-Ž3-pyridyl.-1-butanone, Cancer Res. 47 Ž1987. 1143–1148. w13x S.A. Belinsky, C.M. White, J.A. Boucheron, F.C. Richardson, J.A. Swenberg, M. Anderson, Accumulation and persistence of DNA adducts in respiratory tissue of rats following
w14x
w15x
w16x
w17x
w18x
w19x
w20x
w21x
w22x
w23x
w24x
w25x
multiple administrations of the tobacco specific carcinogen 4-Ž N-methyl-N-nitrosamino.-1-Ž3-pyridyl.-1-butanone, Cancer Res. 46 Ž1986. 1280–1284. P.G. Foiles, N. Trushin, A. Castonguay, Measurement of O 6-methyldeoxyguanosine in DNA methylated by the tobacco-specific carcinogen 4-Žmethylnitrosamino.-1-Ž3-pyridyl.-1-butanone using a biotin–avidin enzyme-linked immunosorbent assay, Carcinogenesis 6 Ž1985. 989–993. V.L. Souliotis, S. Chhabra, L.M. Anderson, S.A. Kyrtopoulos, Dosimetry of O 6-methylguanine in rat DNA after lowdose, chronic exposure to N-nitrosodimethylamine ŽNDMA.. Implications for the mechanism of NDMA hepatocarcinogenesis, Carcinogenesis 16 Ž1995. 2381–2387. C.d. Lindamood, M.A. Bedell, K.C. Billings, M.C. Dyroff, J.A. Swenberg, Dose response for DNA alkylation, w3 H xthymidine uptake into DNA, and O 6 -methylguanineDNA methyltransferase activity in hepatocytes of rats and mice continuously exposed to dimethylnitrosamine, Cancer Res. 44 Ž1984. 196–200. C. Valavanis, V.L. Souliotis, S.A. Kyrtopoulos, Differential effects of procarbazine and methylnitrosourea on the accumulation of O 6-methylguanine and the depletion and recovery of O 6-alkylguanine-DNA alkyltransferase in rat tissues, Carcinogenesis 15 Ž1994. 1681–1688. V.L. Souliotis, C. Valavanis, V.A. Boussiotis, G.A. Pangalis, S.A. Kyrtopoulos, Comparative study of the formation and repair of O 6-methylguanine in humans and rodents treated with dacarbazine, Carcinogenesis 17 Ž1996. 725–732. J.G. Lewis, J.A. Swenberg, The kinetics of DNA alkylation, repair and replication in hepatocytes, Kupffer cells, and sinusoidal endothelial cells in rat liver during continuous exposure to 1,2-dimethylhydrazine, Carcinogenesis 4 Ž1983. 529–536. J.A. Boucheron, F.C. Richardson, P.H. Morgan, J.A. Swenberg, Molecular dosimetry of O 4-ethyldeoxythymidine in rats continuously exposed to diethylnitrosamine, Cancer Res. 47 Ž1987. 1577–1581. V.E. Walker, T.R. Fennell, P.B. Upton, T.R. Skopek, V. Prevost, D.E. Shuker, J.A. Swenberg, Molecular dosimetry of ethylene oxide: formation and persistence of 7-Ž2hydroxyethyl.guanine in DNA following repeated exposures of rats and mice, Cancer Res. 52 Ž1992. 4328–4334. P. Buss, M. Caviezel, W.K. Lutz, Linear dose–response relationship for DNA adducts in rat liver from chronic exposure to aflatoxin B1, Carcinogenesis 11 Ž1990. 2133– 2135. C.P. Wild, R.C. Garner, R. Montesano, F. Tursi, Aflatoxin B1 binding to plasma albumin and liver DNA upon chronic administration to rats, Carcinogenesis 7 Ž1986. 853–858. M.C. Poirier, N.F. Fullerton, B.A. Smith, F.A. Beland, DNA adduct formation and tumorigenesis in mice during the chronic administration of 4-aminobiphenyl at multiple dose levels, Carcinogenesis 16 Ž1995. 2917–2921. C.E. Frantz, C. Bangerter, E. Fultz, K.M. Mayer, J.S. Vogel, K.W. Turteltaub, Dose-response studies of MeIQx in rat liver and liver DNA at low doses, Carcinogenesis 16 Ž1995. 367–373.
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247 w26x K. Yamashita, M. Adachi, S. Kato, H. Nakagama, M. Ochiai, K. Wakabayashi, S. Sato, M. Nagao, T. Sugimura, DNA adducts formed by 2-amino-3,8-dimethylimidazow4,5-fxquinoxaline in rat liver: dose-response on chronic administration, Jpn. J. Cancer Res. 81 Ž1990. 470–476. w27x H.A. Schut, C.R. Herzog, D.A. Cummings, Accumulation of DNA adducts of 2-amino-3-methylimidazow4,5-fx quinoline ŽIQ. in tissues and white blood cells of the Fischer-344 rat after multiple oral dosing, Carcinogenesis 15 Ž1994. 1467– 1470. w28x C.D. Davis, H.A. Schut, E.G. Snyderwine, Adduction of the heterocyclic amine food mutagens IQ and PhIP to mitochondrial and nuclear DNA in the liver of Fischer-344 rats, Carcinogenesis 15 Ž1994. 641–645. w29x D.K. La, J.R. Froines, Formation and removal of DNA adducts in Fischer-344 rats exposed to 2,4-diaminotoluene, Arch. Toxicol. 69 Ž1994. 8–13. w30x M.C. Poirier, F.A. Beland, F.H. Deal, J.A. Swenberg, DNA adduct formation and removal in specific liver cell populations during chronic dietary administration of 2-acetylaminofluorene, Carcinogenesis 10 Ž1989. 1143–1145. w31x M.C. Poirier, B. True, B.A. Laishes, Formation and removal of Žguan-8-yl.-DNA-2-acetylaminofluorene adducts in liver and kidney of male rats given dietary 2-acetylaminofluorene, Cancer Res. 42 Ž1982. 1317–1321. w32x M.C. Poirier, N.F. Fullerton, T. Kinouchi, B.A. Smith, F.A. Beland, Comparison between DNA adduct formation and
w33x
w34x
w35x
w36x
w37x
w38x
247
tumorigenesis in livers and bladders of mice chronically fed 2-acetylaminofluorene, Carcinogenesis 12 Ž1991. 895–900. R.C. Gupta, K. Earley, N.F. Fullerton, F.A. Beland, Formation and removal of DNA adducts in target and nontarget tissues of rats administered multiple doses of 2-acetylaminophenanthrene, Carcinogenesis 10 Ž1989. 2025–2033. D.L. Tullis, K.L. Dooley, D.W. Miller, K.P. Baetcke, F.F. Kadlubar, Characterization and properties of the DNA adducts formed from N-methyl-4-aminoazobenzene in rats during a carcinogenic treatment regimen, Carcinogenesis 8 Ž1987. 577–583. X.S. Cui, U.B. Torndal, L.C. Eriksson, L. Moller, Early ¨ formation of DNA adducts compared with tumor formation in a long-term tumor study in rats after administration of 2-nitrofluorene, Carcinogenesis 16 Ž1995. 2135–2141. J.M. Arif, R.C. Gupta, Detection of DNA-reactive metabolites in serum and their tissue distribution in mice exposed to multiple doses of carcinogen mixtures: role in human biomonitoring, Carcinogenesis 17 Ž1996. 2213–2219. N.J. Gorelick, D.A. Hutchins, S.R. Tannenbaum, G.N. Wogan, Formation of DNA and hemoglobin adducts of fluoranthene after single and multiple exposures, Carcinogenesis 10 Ž1989. 1579–1587. S. Werner, J. Topinka, T. Wolff, L.R. Schwarz, Accumulation and persistence of DNA adducts of the synthetic steroid cyproterone acetate in rat liver, Carcinogenesis 16 Ž1995. 2369–2372.
Correlation of DNA adduct levels with tumor incidence: carcinogenic potency of DNA adducts Michael Otteneder, Werner K. Lutz
)
Department of Toxicology, UniÕersity of Wuerzburg, Versbacher Strasse 9, D-97078 Wuerzburg, Germany Accepted 11 October 1998
Abstract The quantitative relationship between DNA adducts and tumor incidence is evaluated in this review. All available data on DNA adduct levels determined after repeated administration of a carcinogen to rats or mice have been compiled. The list comprised 27 chemicals, of all major structural classes of carcinogens. For the correlation with tumor incidence, the DNA adduct levels measured at the given dose were normalized to the dose which resulted in a 50% tumor incidence under the conditions of a 2-year bioassay ŽTD50 dose.. In rat liver, the calculated adduct concentration ‘responsible’ for a 50% hepatocellular tumor incidence spanned from 53 to 2083 adducts per 10 8 nucleotides, for aflatoxin B1, tamoxifen, IQ, MeIQx, 2,4-diaminotoluene, and dimethylnitrosamine Žin this order.. In mouse liver, the respective figures were 812 to 5543 adducts per 10 8 nucleotides, for ethylene oxide, dimethylnitrosamine, 4-aminobiphenyl, and 2-acetylaminofluorene. The observed span Ž40-fold in rats, 7-fold in mice. reflects differences between the various DNA adducts to lead to critical mutations. If additional carcinogens fit in with this astonishingly narrow range, the measurement of DNA adduct levels in target tissue has the potential to be not only an exposure marker but an individual cancer risk marker. For toremifen and styrene, low levels of DNA adducts were detected in rat liver at the end of a negative long-term bioassay. This shows that the limit of detection of DNA adducts can be well below the limit of detection of an increased tumor incidence. For a cancer risk assessment at low levels of DNA damage, treatment-related adducts must be discussed in relation to the background DNA damage and its inter- and intraindividual variability. q 1999 Elsevier Science B.V. All rights reserved. Keywords: Cancer risk; Carcinogen; DNA adduct; Dose–response relationship; Neoplasm
1. Introduction DNA adducts are constantly being formed, even in the absence of exposure to genotoxic carcinogens X
Abbreviations: dG, 2 -Deoxyguanosine; MeIQx, 2-Amino3,8-dimethyl-imidazow4,5-fxquinoxaline; NNK, 4-Ž N-Methyl-NX nitrosamino.-1-Ž3-pyridyl.-1-butanone; 8-oxodG, 8-Oxo-2 -deoxyguanosine ) Corresponding author. Tel.: q49-931-201 5402; Fax: q49931-201 3446; E-mail: [email protected]
ŽRef. w1x, and references therein.. Humans get cancer even if they avoid exposure to the established cancer risk factors. Similarly, laboratory animals kept under well-controlled ‘carcinogen-free’ conditions show a spontaneous tumor incidence w2x. The question therefore is whether, and to what extent, background DNA adducts could be responsible for the spontaneously arising tumors. The efficiency of the ‘background DNA damage’ to induce ‘spontaneous tumors’ is unknown. The relationship between DNA adduct levels and tumor
0027-5107r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved. PII: S 0 0 2 7 - 5 1 0 7 Ž 9 9 . 0 0 0 2 2 - 6
238
Table 1 DNA adduct levels determined after repeated administration of carcinogens Carcinogen
Speciesr Sex
(A) Methylating and ethylating carcinogens N-Methyl-N-nitrosourea ŽMNU. Rm
Route
Dose Žmg kgy1 dy1 .
Duration
Site
Adduct concentration Žper 10 8 Nt.
Adduct types
Method
Reference
f34
wat
33.3 a
2w
640 1300 4100 46 1016 724 407 44 180 550 840 2500 175000 60000 250 7 10 38r3880 –r24120 120r35970 94r6900 1940r10000 190r9300 875r10200 2660r24800 8r28r78 3r13r39
dG-O 6-Me
Immunoassay
w10x
dG-O 6 -Me
Immunoassay
w11x
dG-O 6 -Me
HPLCrfluorescence and immunoassay
w12x
dG-O 6 -Mer dG-N 7 -Mer dT-O 4-Me dG-O 6 -Me dG-O 6 -Me
HPLCrfluorescence and immunoassay
w13x
Immunoassay Repair assayr ada protein HPLCrfluorescence
w14x w15x
N-Methyl-N -nitro-Nnitrosoguanidine ŽMNNG.
Rm
wis
wat
3.3 a
12 w
4-Ž N-Methyl-N-nitrosamino.-1Ž3-pyridyl.-1-butanone ŽNNK.
Rm
f34
ipj
12 d
Rm
f34
ipj
0.3 1 10 30 100
liv col for for stg duo jej lun
12 d
liv
Rm Rf
f34 wis
ipj wat
14 d 28 d
lun liv
Rm
f34
wat
16 d
liv
Mm
c3h
wat
16 d
liv
Mm
cbl
wat
16 d
liv
Procarbazine
Rf
sda
eat
10 d
liv lyd
Dacarbazine
Rf
sda
ipj
40 0.223 a 0.372 a 0.7 1.9 3.9 1.9 a 4.8a 1.8 4.8 9.5 5 10 20 20
10 d
1,2-Dimethylhydrazine ŽSDMH. Diethylnitrosamine ŽDEN.
Rm Rm
f34 f34
wat wat
2.2 0.03 a 0.08 0.3 0.8 3.3 8.3
16 d 70 d
liv lyd liv liv
X
Dimethylnitrosamine ŽDMN.
84 39 11300 3 3 42 112 224 224
dG-O 6-Mer dG-N 7 -Me
w16x
dG-O 6 -Mer dG-N 7 -Me dG-O 6 -Mer dG-N 7 -Me
HPLCr fluorescence w16x
dG-O 6 -Me
Repair assayr ada protein
dG-O 6 -Me
w18x Repair assayr ada protein HPLCr fluorescence w19x w20x Immunoassay
dG-O 6 -Mer dG-N 7 -Me
HPLCr fluorescence w16x
w17x
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
Strain
Rf
f34
eat
Toremifen ŽTOR.
Rf
f34
eat
Ethylene oxide ŽETO.
Mm
b6c
inh
Aflatoxin B1 ŽAFB1.
Rm
f34
wat
Styrene
Rm R mqf
wis cdr
2-Amino-3-methylimidazow4,5-fxquinoline ŽIQ.
300 610 7 13 7 1016 718
Six addrRAL
Postlab.
w4x
Six addrRAL
Postlab.
w4x
dG-N 7 -HOEt
HPLCr fluorescence
w21x
liv
0.1 3.2 85
Total radioact.
3
w22x
23 d 2y
liv liv
20 90
Total radioact. dG-O 6 -SO
3
H postlab.
w23x unpubl.
28 d
liv
dG-C8-ABP
postlab.
w24x
28 d
liv
dG-C8-ABP
postlab.
w24x
42 d
liv
dG-C8-MeIQx
liv
RAL, 7 spots
accel. mass spectrometry postlab.
w25x
12 w
w26x
15 d
liv sto liv sto liv
1170 4880 7470 9100 220 1220 2520 4400 0.0019 0.5 1 10 100 1000 200 50 750 100 500
RAL, add.stand.
postlab.
w27x
RAL, add.stand.
postlab.
w28x
18 m
liv
18 m
liv
28 d Ž5 drw.
lun liv
2.2=10y6 73=10y6 y3 2.1=10
8w
gav inh
1=10y3 1030
bal
wat
Mm
bal
wat
Rm
sda
eat
Rm
f34
eat
Rm
f34
gav
2.7 10.7 21.4 42.9 2a 7.8 15.7 31.4 9.5=10y6 5=10y3 0.007 0.065 0.672 6.3 5
(C) Aromatic amines and nitroarenes 4-Aminobiphenyl Mf Ž4-ABP.
2-Amino-3,8-dimethylimidazow4,5-fxquinoxaline ŽMeIQx.
a
20.8 41.7 20.8 a 41.7 62.5 b 67
a
50 Rm
f34
gav
100
10 d
HrHPLC
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
(B) Adduct formation Õia epoxide Tamoxifen ŽTAM.
239
240
Table 1 Žcontinued. Carcinogen
Speciesr Sex
Route
Dose Žmg kgy1 dy1 .
Duration
Site
Adduct concentration Žper 10 8 Nt.
Adduct types
Method
Reference
f34
gav
100
10 d
liv
50
RAL, add.stand.
postlab.
w28x
f34 f34 wf
ipj eat eat
postlab. immunoassay immunoassay
w29x w30x w31x
eat
dG-C8-AF
postlab.
w32x
2-Acetylaminophenanthrene Ž2-AAP.
R mqf
sda
ipj
4w
w33x
Rm
cdr
eat
50 a
5w
3
w34x
2-Nitrofluorene ŽNF.
Rm
wis
eat
33
10 d
dG-C8-AP dG-N 2 -AP 2 spots dG-C8-MAB dG-N 2 -MAB dA-N 6 -MAB dG-C8-AF
postlab.
N-Methyl-4-aminoazobenzene ŽMAB.
liv kid ubl liv spl kid liv kid for
330 7200 7400 7900 153r353 563r816 909r2094 1228r3425 1081r3750 1616r4675 1934r6831 2690r8966 200 122 131 900 150 50 11 6 54
3 spots, RAL dG-C8-AF dG-C8-AF
bal
10 d 28 d 31 d 46 d 28 d
liv liv liv
Mf
5 16.7 a 16.7 a 33.3 0.7 a 2.1 4.3 6.4 8.6 10.7 14.2 21.4 0.7 Ž1= wk.
(D) Polynuclear aromatic hydrocarbons Benzow axpyrene ŽBP. Mm
cbl
ipj
1
10 d
Benzow k xfluoranthene ŽBF.
Mm
cbl
ipj
1
10 d
Cyclopentaw cd xpyrene ŽCPP.
Mm
cbl
ipj
1
10 d
Fluoranthene ŽFA.
Rm
sda
eat
20.8 Ž13d. 41.7 Ž24d. mean: 27 a
37 d Ž13q24.
liv lun liv lun liv lun liv kid lun
3 2 3 8 12 36 49 48
Rf Rm
wis wis
gav gav
0.05 0.05
42 d 42 d
liv liv
36.5 0.6
(C) Aromatic amines and nitroarenes 2-Amino-1-methyl-6-phenylimidazo R m w4,5-fxpyrimidine ŽPhIP. 2,4-Diaminotoluene ŽDAT. R n.d. 2-Acetylaminofluorene Ž2-AAF. Rm Rm
(E) Miscellaneous Cyproterone acetate ŽCPA.
livrubl
HrHPLC
postlab.
w35x
RAL
postlab.
w36x
RAL
postlab.
w36x
RAL
postlab.
w36x
dN-anti-FADE, RAL
postlab.
w37x
4 spots, RAL 4 spots, RAL
postlab. postlab.
w38x
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
Strain
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
incidence could be investigated in experiments involving chronic exposure to genotoxic chemical carcinogens, such as the 2-year rodent bioassay for carcinogenicity. If a strong quantitative association can be shown for exogenous carcinogens, the observed levels of background adducts could be correlated with the control tumor incidence. The data available so far are compiled, analyzed, and discussed in this paper. 2. DNA adduct levels after repeated administration of carcinogens
241
analysis of a correlation between adduct levels and tumor incidence because the relationship will strongly depend on the time of analysis after the administration. The best data could be obtained if adduct levels were determined at the end of a standard bioassay for carcinogenicity. Unfortunately, such data are usually not available, with the exception of tamoxifen and toremifen Ž18-month treatment. w4x and the unpublished data for styrene mentioned below Ž2-year treatment.. As a compromise, DNA adduct levels reached after repeated administration for at least 10 days were included in the analysis.
2.1. DNA-adduct leÕels as a function of time
2.2. Literature search for DNA adduct data after repeated administration of carcinogens
The concentration of adducts measured in a DNA sample is the result of a balance between rates of formation, removal by repair and cell death, and ‘dilution’ by DNA replication Žthe newly synthesized strand does not contain adducts.. After a long period of treatment, a steady state is expected to be reached, except for repair-resistant adducts in non-dividing cells. For adducts that are rapidly repaired, steadystate levels are dominated by repair; for adducts that are refractory to repair, the cell turnover will be the rate-limiting factor. Most published DNA-adduct data have been determined after a single dose of a carcinogen w3x. These data are of limited value for a quantitative
Two Searches in Medline yielded 344 and 378 hits, of which 30 references contained valuable data on 34 experiments. Adduct levels derived from dermal administration were not included. A minimum duration of treatment for 10 days was required for the reference to be evaluated. Table 1 summarizes the results. Data were available for carcinogens from a variety of classes, 7 methylating agents and 1 ethylating agent Žpart A., 5 agents which alkylate via an epoxide Žpart B., 7 aromatic amines and 1 nitroarene Žpart C., 4 polynuclear aromatic hydrocarbon Žpart D., and 1 steroid analogue Žpart E.. Most data were available for the liver. Other tissues such as lung, kidney and
Notes to Table 1: Species: M, mouse; R, rat. Sex: f, female; m, male; n.d., no data. Strain: bal, BALBrc; b6c, B6C3F1 ; cbl, C57BL; c3h, C3H; cdr, CD Charles River; f34, F344; sda, Sprague–Dawley; wf, WF; wis, Wistar. Route: eat, diet; gav, gavage; inh, by inhalation; ipj, intraperitoneal injection; wat, in drinking water. Dose: a The concentration in the drinking water or in the feed Žexpressed in ppm. was converted to mg kgy1 dy1 by division by 12 Žrat. or 7 Žmouse.. b The concentration in the air Žexpressed in ppm. was converted to mg kgy1 dy1 on the basis of a ventilation rate of 0.03 l mousey1 miny1 Ž0.25 lrrat. and 100% absorption. The underlined values were used for the calculation of adduct levels in Table 2. Duration: d, day; w, week; m, month; y, year. Site: col, colon; duo, duodenum; for, forestomach; jej, jejunum; kid, kidney; liv, liver; lun, lung; lyd, lymph node; sto, stomach; stg, stomach, glandular; ubl, urinary bladder; Adduct concentration: Nt, nucleotide; conversion of units expressed per parent base was based on a molecular percentage of 22% G and C, 28% A and T. The underlined values were used for the calculation of adduct levels in Table 2. Adduct type: add, adduct; dA-N 6 , N 6-alkylated deoxyadenosine; dG-O 6 , O 6-alkylated deoxyguanosine; dG-N 2 , N 2-alkylated deoxyguanosine, dG-C8, C8-alkylated deoxyguanosine; dG-N 7, N 7-alkylated deoxyguanosine; dT-O 4 , O 4-alkylated deoxythymidine; Et, Ethyl; FADE, 2,3-dihydroxy-1,10b-epoxy-1,2,3-trihydro-fluoranthene; Me, Methyl; RAL, relative adduct labeling; SO, styrene oxide; total, total measured DNA-adduct level. Method: accel. mass spectrometry, accelerator mass spectrometry; 3 H, administration of 3 H-radiolabeled carcinogen; 3 H rHPLC, nucleotide analysis by HPLC after administration of 3 H-radiolabeled carcinogen; postlab., 32 P-postlabeling; repair assayrada protein, competitive repair assay using E. coli O 6-methylguanine–DNA alkyltransferase.
242
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
forestomach were investigated occasionally, depending on the organotropism for tumor induction by the carcinogen in question. Reported DNA adduct levels spanned one hundred million-fold from as low as about 2 adducts per 10 11 rat liver nucleotides, measured 42 days after a daily oral dose of 9.5 ng MeIQx per kg body weight Žadducts analyzed by accelerator mass spectrometry. to as many as 2 adducts per 1000 nucleotides in rat liver DNA, measured after 12 days of 100 mg NNK per kg body weight Žadduct analysis by HPLCrfluorescence.. DNA-reactive carcinogens usually result in multiple adducts. The number can span from a few Že.g., aromatic amines. to more than a dozen Že.g., small methylating agents such as the methyl diazonium ion CH 3 –Nq' N derived from dimethylnitrosamine.. Different adducts can have different consequences. For instance, while the methylation of the nitrogen atom a7 of guanine facilitates depurination, the methylation of guanine at the O 6 results in mispairing with T, thereby producing a G–C to A–T transition mutation after 2 rounds of DNA replication. Methylating agents which produce different ratios of N 7- vs. O 6- methylation therefore are expected to have different effects, possibly also with respect to tumor induction. The number of detected adducts depends on the method used. If the analysis is based on the determination of radioactivity after the administration of appropriately radiolabeled test chemicals, all adducts formed might be detected Že.g., aflatoxin B1.. On the other hand, if the analysis is based on photometric or mass spectrometric detection, it will be adductspecific. With the 32 P-postlabeling method, a quantification on the basis of RAL values Žrelative adduct labeling. could result in an underestimation of the actual adduct levels. These limitations are expected to affect the attempted quantitative correlation of adduct levels with tumor incidence data. 3. DNA adduct levels associated with a 50% liver tumor incidence 3.1. Normalization of adduct leÕels to the tumorigenic dose 50% (TD50 ) If all adducts had the same carcinogenic potency, the same adduct levels would be required for a given
tumor incidence. In order to test this hypothesis, the adduct levels compiled in Table 1 were normalized to a standard dose with respect to the induced tumor incidence. The TD50 value was chosen for this purpose. This is the dose which halves the probability to stay tumor-free within two years, or approximately, it is the dose which results in a 50% tumor incidence within 2 years. Criteria to calculate TD50 values have been set by Lois Gold and coworkers and all values have been published in the carcinogenic potency database which is available in print w5x and on the Internet Žhttp:rrpotency.berkeley.edurcpdb.. A number of criteria had to be met for a carcinogen and a study listed in Table 1 to be included for further evaluation. First, the dose used for the DNAadduct experiment should not be more than 10-fold higher than the TD50 value. When different dose levels were used for the DNA adduct determinations, the dose which was closest to the TD50 was used. Second, only those situations were included in the evaluation where the major or the majority of the adducts were analyzed. Third, in view of the scarcity of adduct data for organs other than the liver, only this organ was considered. With respect to histopathological criteria, the TD50 value referring to hepatocellular tumors was used, if available. Published TD50 values were available for most of the chemicals which met these criteria. For tamoxifen citrate, an unpublished value was communicated by Dr. Lois Gold. 3.2. Adducts Õs. tumor incidence The adduct concentrations calculated for the dose levels associated with a 50% liver tumor incidence are shown in Table 2. Proportionality was used for the adjustment, i.e., adduct concentrations given in column a8 of Table 1 were divided by the dose shown in column a5 and multiplied with the TD50 value shown in column a4 of Table 2. The result is shown in column a6 of Table 2. In the rat Žpart A., values span from 53 adducts per 10 8 nucleotides for the aflatoxin B1 to 2082 per 10 8 nucleotides for dimethylnitrosamine. The data could be interpreted to mean that the aflatoxin–DNA adducts are about 40 times more potent for the induction of hepatocellular carcinoma than the adducts measured with dimethylnitrosamine. In view
Table 2 DNA adduct concentrations in the liver calculated for the dose which results in a 50% hepatic tumor incidence ŽTD50 . in rat ŽA. and mouse ŽB.
(A) Rat Dimethylnitrosamine ŽDMN. Tamoxifen ŽTAM. Aflatoxin B1 ŽAFB1. 2-Amino-3,8-dimethyl-imidazow4,5-fxquinoxaline ŽMeIQx. 2-Amino-3-methylimidazow4.5fxquinoline ŽIQ. 2,4-Diaminotoluene ŽDAT. B. Mouse Dimethylnitrosamine ŽDMN. Ethylene oxide ŽETO. 4-Aminobiphenyl Ž4-ABP. 2-Acetylaminofluorene Ž2-AAF.
Sex
Strain
TD50 Žmg kgy1 dy1 .
Histopathology
Adduct concentration at TD50 Žper 10 8 Nt.
Comments
m f m m
f34 n.d.Ž ™f34. f34 f34
0.372 5.54 0.0013 1.26
bht Ž?. hpc hpc
2082 80 53 188
Major adducts RAL Total adduct level RAL
m
f34
2.08
hpc
83
RAL
fqm Ž ™ n.d..
f34
14
MXA q MXB
924
RAL
m f Ž ™ m. f m f
cbl b6c cif Ž ™ bal. cif Ž ™ bal. bal
0.179 75.8 3.05 7.5 44.1
mal hpa mhp mhp hpc
944 812 1322 1173 5543
Major adducts Single major adduct Single major adduct Single major adduct Single major adduct
Histopathology: bht, benign hepatocellular tumor; hpa: hepatocellular adenoma; hpc: hepatocellular carcinoma; hpt: hepatoma; mal: malignant tumor; mhp: malignant hepatoma; MXA: more than one tumor type, combined by NCIrNTP; MXB: more than one tumor type, combined by the authors of the CPDB at Berkeley. SexrStrain: The information given in the brackets refer to the DNA-adduct data.
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
Carcinogen
243
244
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
of the enormous span of DNA adduct levels compiled in Table 1, this is an astonishingly small span. Within a class of structurally similar carcinogens, for instance aromatic amines, the range is even smaller. It ranges from 83 to 924 per 10 8 nucleotides, i.e., spans only one order of magnitude. This means that adduct levels measured in the liver might indeed have some potential to predict tumor incidence. In the mouse Žpart B of Table 2., only 4 data sets fulfilled the criteria for analysis. The values were higher than in the rat and ranged from 812 to 5543 per 10 8 nucleotides, i.e., spanned only seven-fold. Whether the mouse liver can indeed tolerate larger levels of DNA adducts than the rat without tumor response is an interesting question. Spontaneous liver tumor formation normally is more pronounced in mice, as compared with rats. Although this is strainand sex-dependent, it could indicate that nongenotoxic aspects of carcinogenesis in the mouse liver will have to be given particular attention. 4. Discussion 4.1. Carcinogenic potencies of DNA adducts Based on the data available so far, the range of carcinogenic potency of structurally different DNA adducts is less than 2 orders of magnitude. Within a class of chemically related carcinogens, it is only 1 order of magnitude. This result refers to the liver and the induction of hepatocellular tumors. More data will have to be generated before the same analysis can be performed for other organs. Furthermore, the data are expected to be cell type-specific. In the liver, about 70% of the cells are hepatocytes. DNA isolated from a homogenate of this organ therefore is predominantly from the target cell type. In organs such as the lung or the kidney, an average level of DNA adducts determined in an organ homogenate is unlikely to represent the situation prevailing in a specific target cell population. For such situations, adduct data will have to be collected from enriched cell populations. 4.2. Measurable adduct leÕels in the absence of increased tumor incidence (example: styrene) The correlation established between adduct levels and tumor incidence also allows to discuss the ques-
tion whether an adduct concentration might be measurable without giving rise to a detectable increase in tumor incidence. With 50 animals per group as used in a standard bioassay for carcinogenicity and assuming a 5% background tumor incidence in the controls, a 5% increase in a dose group would not be statistically significantly different from the controls. A 5% increase is one-tenth of the incidence used to normalize the adduct levels. Therefore, an adduct level of up to 200 per 10 8 nucleotides in the rat liver and up to 500 per 10 8 nucleotides in mouse liver could go unobserved with respect to an increased tumor incidence. This upper limit is compatible with unpublished data obtained from liver DNA of rats treated for 2 years by inhalation with styrene. The main intermediate metabolite of styrene is styrene-7,8-oxide. This epoxide has weakly electrophilic activity and has a plasma half life of about 20 min in rats. Low levels of DNA adducts are therefore expected in all tissues. In some qualitative contrast, there was no evidence of carcinogenic activity in a recent 2-year bioassay for carcinogenicity of styrene by inhalation exposure of CD-rats to 50, 200, 500 and 1000 ppm, for 6 hrday and 5 daysrweek. Liver tissue was available from the rats at the end of the 2-year exposure. The 32 P-postlabeling technique was optimized for the detection and quantification of the 2 regioisomeric 2X-deoxyguanosyl-O 6-adducts from styrene-7,8-oxide w6x, and the liver DNA was analyzed. In the 1000 ppm group, total adduct levels were about 100 adducts per 10 8 nucleotides. This level is below the value mentioned above, expected to result in a statistically significant increase in tumor incidence. It shows that the limit of detection of DNA adduct levels can be much lower than the limit of detection for a significant increase in tumor incidence. In this example, the 32 P-postlabeling technique was used. Under optimal conditions, limits of detection of one adduct per 10 9 nucleotides can be achieved. This is equivalent to 20 adducts per diploid cellular genome containing 6 pg DNA. Using accelerator mass spectrometry, the adduct levels determined after MeIQx exposure corresponded to not even 1 adduct per cell. In view of the background DNA damage present at all time, it goes without saying that some adduct levels associated with an external exposure can be mea-
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
sured but cannot constitute a cancer risk worth mentioning. 4.3. Background DNA damage and spontaneous tumor incidence The question shall now be addressed whether the background DNA adduct level determined in the rat liver is compatible with the observed spontaneous liver tumor incidence. In most rat strains, the latter is in the low percent range. Assuming a spontaneous tumor incidence of 5%, an upper limit adduct level of 200 per 10 8 nucleotides could be associated with this tumor incidence. For I-compound adducts and for adducts derived from alkylating agents, aldehydes, and aromatic amines, this number would probably not be exceeded w1x, indicating that the background DNA adducts do not differ from the adducts formed by the exogenous chemical carcinogens with respect to carcinogenic potency. For DNA oxidations, the situation is critically dependent on the choice of reference for the levels of 8-oxo-deoxyguanosine and on the question of the levels of other oxidation products. On the basis of the recently published values for rat liver Ž0.23 8-oxodG per 10 5 dG, i.e., 50 8-oxodG per 10 8 nucleotides w7x., the level of 200 damaged nucleotides per 10 8 nucleotides would not be exceeded. However, taking into account that 8-oxodG is only one of a number of oxidative DNA lesions, the data could be also interpreted to mean that the oxidative background DNA damage is less dangerous than other types of DNA damage. Evolutionary pressure against oxygen-related damage might have resulted in the formation of specific, replication-coupled repair systems for oxidative DNA lesions. 4.4. Correlation between DNA adduct leÕels and cancer risk The use of DNA adduct data from chronic carcinogen exposure for cancer risk assessment has been discussed before by Poirier and Beland w8x. These authors focused on the shape of the dose-response curves for DNA adduct levels and tumor incidence in various organs, and compared the respective data for animals with humans. As an exam-
245
ple, the relationship between DNA adduct levels and bladder tumor incidence induced in male mice by 4-aminobiphenyl was compared with the situation in human smokers. The authors concluded that, per unit dose, humans may be more sensitive than mice to tumor induction by aromatic amine exposure andror that factors in addition to carcinogen exposure are contributing to the tumorigenic end point. A number of critical assumptions had to be made for this conclusion, in particular with respect to the adduct levels in the human bladder. It is hoped that biomonitoring data will accumulate in the future so that the uncertainties can be reduced. The question of a correlation of a DNA adduct level with tumor incidence on an individual basis has been addressed experimentally. Using the mouse skin tumor model and chronic exposure to 7,12-dimethylbenzw axanthracene ŽDMBA., the individual latency time to the appearance of the first papilloma was correlated with the DMBA–DNA adduct levels in the treated normal skin areas. Surprisingly, the adduct levels were lowest in those mice which showed the shortest latency period w9x. The explanation came from the analysis of the rate of cell division in the epidermis. The mice with the fastest tumor induction had the highest rates of cell division. Since the DNA adduct concentration is reduced by a factor of 2 with each round of DNA replication, cell division must be considered a confounder for the interpretation of DNA adduct levels. Our analysis showed that adducts derived from a variety of genotoxic chemical carcinogens differed by 1–2 orders of magnitude with respect to their carcinogenic potency in rodent liver. This range is small if compared with the range of possible exposures. Therefore, for a specific organ, the determination of a DNA adduct level could become predictive for a cancer risk. What is needed are more data on adduct levels in target organs of animals and humans. This would allow to interpret DNA adduct data not only as exposure markers but individual risk markers.
Acknowledgements We thank Dr. Lois Gold for making unpublished TD50 values available. Liver samples of styrene-
246
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247
treated rats were a gift of the Styrene Information and Research Center ŽSIRC.. The styrene adduct analysis was supported by the Styrene Steering Committee of CEFIC.
References w1x R.C. Gupta, W.K. Lutz, Background DNA damage from endogenous and unavoidable exogenous exposure to genotoxic carcinogens, Mutat. Res. Ž1999., in press. w2x J.K. Haseman, J. Huff, G.A. Boorman, Use of historical control data in carcinogenicity studies in rodents, Toxicol. Pathol. 12 Ž1984. 126–135. w3x W.K. Lutz, In vivo covalent binding of organic chemicals to DNA as a quantitative indicator in the process of chemical carcinogenesis, Mutat. Res. 65 Ž1979. 289–356. w4x D. Li, Y. Dragan, V.C. Jordan, M. Wang, H.C. Pitot, Effects of chronic administration of tamoxifen and toremifene on DNA adducts in rat liver, kidney, and uterus, Cancer Res. 57 Ž1997. 1438–1441. w5x L.S. Gold, E. Zeiger, Handbook of Carcinogenic Potency and Genotoxicity Database, CRC Press, Boca Raton, FL 1997. w6x M. Otteneder, E. Eder, W.K. Lutz, 32 P-postlabeling analysis of DNA adducts of styrene-7,8-oxide at the O 6-position of guanine, Submitted to Chem. Res. Toxicol. 12 Ž1999. 93–99. w7x T. Hofer, L. Moller, Reduction of oxidation during the ¨ X preparation of DNA and analysis of 8-hydroxy-2 -deoxyguanosine, Chem. Res. Toxicol. 11 Ž1998. 882–887. w8x M.C. Poirier, F.A. Beland, DNA adduct measurements and tumor incidence during chronic carcinogen exposure in animal models: implications for DNA adduct-based human cancer risk assessment, Chem. Res. Toxicol. 5 Ž1992. 749–755. w9x W. Fischer, W.K. Lutz, Correlation of individual papilloma X latency time with DNA adducts, 8-hydroxy-2 -deoxyguanosine, and the rate of DNA synthesis in the epidermis of mice treated with 7,12-dimethylbenzwaxanthracene, Proc. Natl. Acad. Sci. U.S.A. 92 Ž1995. 5900–5904. w10x H. Ohgaki, B.I. Ludeke, I. Meier, P. Kleihues, W.K. Lutz, C. Schlatter, DNA methylation in the digestive tract of F344 rats during chronic exposure to N-methyl-N-nitrosourea, J. Cancer Res. Clin. Oncol. 117 Ž1991. 13–18. w11x N.H. Zaidi, C.S. Potten, G.P. Margison, D.P. Cooper, P.J. O’Connor, Tissue and cell specific methylation, repair and synthesis of DNA in the upper gastrointestinal tract of Wistar X rats treated with N-methyl-N -nitro-N-nitrosoguanidine via the drinking water, Carcinogenesis 14 Ž1993. 1991–2001. w12x S.A. Belinsky, C.M. White, T.R. Devereux, J.A. Swenberg, M.W. Anderson, Cell selective alkylation of DNA in rat lung following low dose exposure to the tobacco specific carcinogen 4-Ž N-methyl-N-nitrosamino.-1-Ž3-pyridyl.-1-butanone, Cancer Res. 47 Ž1987. 1143–1148. w13x S.A. Belinsky, C.M. White, J.A. Boucheron, F.C. Richardson, J.A. Swenberg, M. Anderson, Accumulation and persistence of DNA adducts in respiratory tissue of rats following
w14x
w15x
w16x
w17x
w18x
w19x
w20x
w21x
w22x
w23x
w24x
w25x
multiple administrations of the tobacco specific carcinogen 4-Ž N-methyl-N-nitrosamino.-1-Ž3-pyridyl.-1-butanone, Cancer Res. 46 Ž1986. 1280–1284. P.G. Foiles, N. Trushin, A. Castonguay, Measurement of O 6-methyldeoxyguanosine in DNA methylated by the tobacco-specific carcinogen 4-Žmethylnitrosamino.-1-Ž3-pyridyl.-1-butanone using a biotin–avidin enzyme-linked immunosorbent assay, Carcinogenesis 6 Ž1985. 989–993. V.L. Souliotis, S. Chhabra, L.M. Anderson, S.A. Kyrtopoulos, Dosimetry of O 6-methylguanine in rat DNA after lowdose, chronic exposure to N-nitrosodimethylamine ŽNDMA.. Implications for the mechanism of NDMA hepatocarcinogenesis, Carcinogenesis 16 Ž1995. 2381–2387. C.d. Lindamood, M.A. Bedell, K.C. Billings, M.C. Dyroff, J.A. Swenberg, Dose response for DNA alkylation, w3 H xthymidine uptake into DNA, and O 6 -methylguanineDNA methyltransferase activity in hepatocytes of rats and mice continuously exposed to dimethylnitrosamine, Cancer Res. 44 Ž1984. 196–200. C. Valavanis, V.L. Souliotis, S.A. Kyrtopoulos, Differential effects of procarbazine and methylnitrosourea on the accumulation of O 6-methylguanine and the depletion and recovery of O 6-alkylguanine-DNA alkyltransferase in rat tissues, Carcinogenesis 15 Ž1994. 1681–1688. V.L. Souliotis, C. Valavanis, V.A. Boussiotis, G.A. Pangalis, S.A. Kyrtopoulos, Comparative study of the formation and repair of O 6-methylguanine in humans and rodents treated with dacarbazine, Carcinogenesis 17 Ž1996. 725–732. J.G. Lewis, J.A. Swenberg, The kinetics of DNA alkylation, repair and replication in hepatocytes, Kupffer cells, and sinusoidal endothelial cells in rat liver during continuous exposure to 1,2-dimethylhydrazine, Carcinogenesis 4 Ž1983. 529–536. J.A. Boucheron, F.C. Richardson, P.H. Morgan, J.A. Swenberg, Molecular dosimetry of O 4-ethyldeoxythymidine in rats continuously exposed to diethylnitrosamine, Cancer Res. 47 Ž1987. 1577–1581. V.E. Walker, T.R. Fennell, P.B. Upton, T.R. Skopek, V. Prevost, D.E. Shuker, J.A. Swenberg, Molecular dosimetry of ethylene oxide: formation and persistence of 7-Ž2hydroxyethyl.guanine in DNA following repeated exposures of rats and mice, Cancer Res. 52 Ž1992. 4328–4334. P. Buss, M. Caviezel, W.K. Lutz, Linear dose–response relationship for DNA adducts in rat liver from chronic exposure to aflatoxin B1, Carcinogenesis 11 Ž1990. 2133– 2135. C.P. Wild, R.C. Garner, R. Montesano, F. Tursi, Aflatoxin B1 binding to plasma albumin and liver DNA upon chronic administration to rats, Carcinogenesis 7 Ž1986. 853–858. M.C. Poirier, N.F. Fullerton, B.A. Smith, F.A. Beland, DNA adduct formation and tumorigenesis in mice during the chronic administration of 4-aminobiphenyl at multiple dose levels, Carcinogenesis 16 Ž1995. 2917–2921. C.E. Frantz, C. Bangerter, E. Fultz, K.M. Mayer, J.S. Vogel, K.W. Turteltaub, Dose-response studies of MeIQx in rat liver and liver DNA at low doses, Carcinogenesis 16 Ž1995. 367–373.
M. Otteneder, W.K. Lutz r Mutation Research 424 (1999) 237–247 w26x K. Yamashita, M. Adachi, S. Kato, H. Nakagama, M. Ochiai, K. Wakabayashi, S. Sato, M. Nagao, T. Sugimura, DNA adducts formed by 2-amino-3,8-dimethylimidazow4,5-fxquinoxaline in rat liver: dose-response on chronic administration, Jpn. J. Cancer Res. 81 Ž1990. 470–476. w27x H.A. Schut, C.R. Herzog, D.A. Cummings, Accumulation of DNA adducts of 2-amino-3-methylimidazow4,5-fx quinoline ŽIQ. in tissues and white blood cells of the Fischer-344 rat after multiple oral dosing, Carcinogenesis 15 Ž1994. 1467– 1470. w28x C.D. Davis, H.A. Schut, E.G. Snyderwine, Adduction of the heterocyclic amine food mutagens IQ and PhIP to mitochondrial and nuclear DNA in the liver of Fischer-344 rats, Carcinogenesis 15 Ž1994. 641–645. w29x D.K. La, J.R. Froines, Formation and removal of DNA adducts in Fischer-344 rats exposed to 2,4-diaminotoluene, Arch. Toxicol. 69 Ž1994. 8–13. w30x M.C. Poirier, F.A. Beland, F.H. Deal, J.A. Swenberg, DNA adduct formation and removal in specific liver cell populations during chronic dietary administration of 2-acetylaminofluorene, Carcinogenesis 10 Ž1989. 1143–1145. w31x M.C. Poirier, B. True, B.A. Laishes, Formation and removal of Žguan-8-yl.-DNA-2-acetylaminofluorene adducts in liver and kidney of male rats given dietary 2-acetylaminofluorene, Cancer Res. 42 Ž1982. 1317–1321. w32x M.C. Poirier, N.F. Fullerton, T. Kinouchi, B.A. Smith, F.A. Beland, Comparison between DNA adduct formation and
w33x
w34x
w35x
w36x
w37x
w38x
247
tumorigenesis in livers and bladders of mice chronically fed 2-acetylaminofluorene, Carcinogenesis 12 Ž1991. 895–900. R.C. Gupta, K. Earley, N.F. Fullerton, F.A. Beland, Formation and removal of DNA adducts in target and nontarget tissues of rats administered multiple doses of 2-acetylaminophenanthrene, Carcinogenesis 10 Ž1989. 2025–2033. D.L. Tullis, K.L. Dooley, D.W. Miller, K.P. Baetcke, F.F. Kadlubar, Characterization and properties of the DNA adducts formed from N-methyl-4-aminoazobenzene in rats during a carcinogenic treatment regimen, Carcinogenesis 8 Ž1987. 577–583. X.S. Cui, U.B. Torndal, L.C. Eriksson, L. Moller, Early ¨ formation of DNA adducts compared with tumor formation in a long-term tumor study in rats after administration of 2-nitrofluorene, Carcinogenesis 16 Ž1995. 2135–2141. J.M. Arif, R.C. Gupta, Detection of DNA-reactive metabolites in serum and their tissue distribution in mice exposed to multiple doses of carcinogen mixtures: role in human biomonitoring, Carcinogenesis 17 Ž1996. 2213–2219. N.J. Gorelick, D.A. Hutchins, S.R. Tannenbaum, G.N. Wogan, Formation of DNA and hemoglobin adducts of fluoranthene after single and multiple exposures, Carcinogenesis 10 Ž1989. 1579–1587. S. Werner, J. Topinka, T. Wolff, L.R. Schwarz, Accumulation and persistence of DNA adducts of the synthetic steroid cyproterone acetate in rat liver, Carcinogenesis 16 Ž1995. 2369–2372.